Howdy, Stranger!

It looks like you're new here. Sign in or register to get started.

Mo’ money mo’ problems – politicalbetting.com

SystemSystem Posts: 12,159
edited March 24 in General
imageMo’ money mo’ problems – politicalbetting.com

Sunak under growing pressure from Tories to return £10m Hester donation https://t.co/ta00pbwBYE

Read the full story here

«13

Comments

  • TheScreamingEaglesTheScreamingEagles Posts: 119,620
    FPT
    DavidL said:

    Completely O/T and with a hat tip to WingsoverScotland but there is a great piece about the introduction of the new Hate Crime Act in Scotland. This is coming into force on April 1st (seriously) and will introduce locations at which these hate crimes can be reported. One of them is a sex shop in Glasgow (honestly).

    Which generated the following brilliant tweet:

    “I’d like some poppers, that black destroyer 12 inch dildo, nipple clamps, a latex tight fitting bondage all in one suit, that gas mask, those restraints” “will that be all sir?” “No I’d like to report a hate crime. Someone shouted pervert at me before I came in here

    https://wingsoverscotland.com/when-theres-no-one-left-to-fight/#more-141782

    BIB - My inner Finbarr Saunders has re-emerged.

    (Well it has never gone away, my nickname at uni was Finn for that reason.)
  • StuartinromfordStuartinromford Posts: 17,214
    Though a majority of 2019 Conservative voters want the party to keep the cash, so it's probably fine in their eyes.

    And if the Conservative Party has decided that its reputation is worth £10 million...

    ... It's not obvious that they haven't got the better side of the deal right now.
  • MightyAlexMightyAlex Posts: 1,660
    It was a story on PM today. Curious to why journalists aren't asking Hester why he gave £10 million to the Conservatives?
  • ohnotnowohnotnow Posts: 3,775
    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

  • MexicanpeteMexicanpete Posts: 28,368

    It was a story on PM today. Curious to why journalists aren't asking Hester why he gave £10 million to the Conservatives?

    He surely didn't get anything in return except for inch-perfect Conservative Government.
  • ohnotnowohnotnow Posts: 3,775

    It was a story on PM today. Curious to why journalists aren't asking Hester why he gave £10 million to the Conservatives?

    He just really, really loves them. Pure altruism. It's a rare thing these days.

    :: wipes tear from eye at the pure giving ::
  • DavidLDavidL Posts: 53,798
    As I said yesterday, the real problem here is who is going to make a significant donation to any political party and risk being Hestered on the basis of something you allegedly said years before? Even trying to buy a seat in the House of Lords is surely not worth that.

    All political parties will feel the draught of this.
  • FF43FF43 Posts: 17,208
    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?
  • Leon said:

    Leon said:

    AI is just the next step of evolution of computing, it's not a revolution. It's the next tool, no more, no less.

    People have been using computers for decades. Computers have been automating elements for decades. AI is not some big unheard of thing, it's just what we already have but the next step along.

    AI is no more going to eliminate employment than computers or robots have done so.

    I agree with your assessment on the non-elimination of employment.

    Again, just on words, in the Academy, in computer science, we use the term "AI" to describe systems that we have been using for decades. We have AI currently: not the next step, but the present step. There is a journal called Artificial Intelligence, https://www.sciencedirect.com/journal/artificial-intelligence/issues , and that's been publishing since 1970.

    The way those AI systems work has changed, from logic-based systems to machine learning and now to generative AI. Their capabilities have increased. In the last few years, we've seen these leaps forward in a set of methods called generative AI, including generative AI systems for language, called large language models (LLMs). Any software that mimics some aspect of intelligence is called AI, even when it is a very limited intelligence.

    A lot of the discussion in this thread is using "AI" to mean something more, something akin to the term "artificial general intelligence" (AGI), i.e. something that can pass the Turing test. We don't have that now. I suspect we won't have that for awhile, although there have been a number of wild predictions suggesting otherwise.

    AI exists, is well-established and already in your smartphone. AI consciousness or AGI, that's another kettle of fish.


    You really think AI won’t eliminate lots of jobs? That’s brave

    In the last couple of months the bank Klarna started using OpenAI to replace customer service staff

    The result was so profound and “shocking” - their words - they put out a press statement as a kind of warning

    “New York, NY – February 27, 2024 – Klarna today announced its AI assistant powered by OpenAI. Now live globally for 1 month, the numbers speak for themselves:

    The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats

    It is doing the equivalent work of 700 full-time agents

    It is on par with human agents in regard to customer satisfaction score

    It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries

    Customers now resolve their errands in less than 2 mins compared to 11 mins previously

    It’s available in 23 markets, 24/7 and communicates in more than 35 languages

    It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024”

    That’s 700 jobs. In one go. In one modest company

    This is the beginning
    Which is bugger all.

    Its not one modest company, its a global company, and its the kind of crap jobs that are churned routinely, indeed its largely outsourced that kind of job anyway precisely because its so unimportant.
    ohnotnow said:

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.
    He’s not alone. Most of the people on this thread have no idea how quickly AI is progressing in all fields - hence, in part, their foolish opinions
    Or we understand it full well, but its moot.

    AI has been progressing for fifty years.

    Automation has been progressing for centuries.

    Automation, of which AI is another step of evolution, hasn't eliminated jobs net.

    Its eliminated some jobs, but then new ones are created, its evolution not revolution.

    AI is a tool, just like spreadsheets and the internet and a bazillion other things are tools. They're tools that will eliminate some menial jobs and create other ones, just like every other form of automation before it.
  • ohnotnowohnotnow Posts: 3,775
    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    Suggest a low emission zone? We're at that kind of stage...
  • MoonRabbitMoonRabbit Posts: 13,493
    On topic. 44% of Tories 2019 voters, and 32% of all voters don’t find his remarks racist at all? 🫢
  • LeonLeon Posts: 55,235
    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
  • TazTaz Posts: 14,362
    On the subject of AI this is interestimg

    https://x.com/taylor_rosie/status/1767853702742774144?s=61
  • MoonRabbitMoonRabbit Posts: 13,493
    DavidL said:

    As I said yesterday, the real problem here is who is going to make a significant donation to any political party and risk being Hestered on the basis of something you allegedly said years before? Even trying to buy a seat in the House of Lords is surely not worth that.

    All political parties will feel the draught of this.

    Good. What a lot of decent money pissed up walls on a load of old shyster consultants.
  • LeonLeon Posts: 55,235

    Leon said:

    Leon said:

    AI is just the next step of evolution of computing, it's not a revolution. It's the next tool, no more, no less.

    People have been using computers for decades. Computers have been automating elements for decades. AI is not some big unheard of thing, it's just what we already have but the next step along.

    AI is no more going to eliminate employment than computers or robots have done so.

    I agree with your assessment on the non-elimination of employment.

    Again, just on words, in the Academy, in computer science, we use the term "AI" to describe systems that we have been using for decades. We have AI currently: not the next step, but the present step. There is a journal called Artificial Intelligence, https://www.sciencedirect.com/journal/artificial-intelligence/issues , and that's been publishing since 1970.

    The way those AI systems work has changed, from logic-based systems to machine learning and now to generative AI. Their capabilities have increased. In the last few years, we've seen these leaps forward in a set of methods called generative AI, including generative AI systems for language, called large language models (LLMs). Any software that mimics some aspect of intelligence is called AI, even when it is a very limited intelligence.

    A lot of the discussion in this thread is using "AI" to mean something more, something akin to the term "artificial general intelligence" (AGI), i.e. something that can pass the Turing test. We don't have that now. I suspect we won't have that for awhile, although there have been a number of wild predictions suggesting otherwise.

    AI exists, is well-established and already in your smartphone. AI consciousness or AGI, that's another kettle of fish.


    You really think AI won’t eliminate lots of jobs? That’s brave

    In the last couple of months the bank Klarna started using OpenAI to replace customer service staff

    The result was so profound and “shocking” - their words - they put out a press statement as a kind of warning

    “New York, NY – February 27, 2024 – Klarna today announced its AI assistant powered by OpenAI. Now live globally for 1 month, the numbers speak for themselves:

    The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats

    It is doing the equivalent work of 700 full-time agents

    It is on par with human agents in regard to customer satisfaction score

    It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries

    Customers now resolve their errands in less than 2 mins compared to 11 mins previously

    It’s available in 23 markets, 24/7 and communicates in more than 35 languages

    It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024”

    That’s 700 jobs. In one go. In one modest company

    This is the beginning
    Which is bugger all.

    Its not one modest company, its a global company, and its the kind of crap jobs that are churned routinely, indeed its largely outsourced that kind of job anyway precisely because its so unimportant.
    ohnotnow said:

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.
    He’s not alone. Most of the people on this thread have no idea how quickly AI is progressing in all fields - hence, in part, their foolish opinions
    Or we understand it full well, but its moot.

    AI has been progressing for fifty years.

    Automation has been progressing for centuries.

    Automation, of which AI is another step of evolution, hasn't eliminated jobs net.

    Its eliminated some jobs, but then new ones are created, its evolution not revolution.

    AI is a tool, just like spreadsheets and the internet and a bazillion other things are tools. They're tools that will eliminate some menial jobs and create other ones, just like every other form of automation before it.
    AI is not “a tool just like a spreadsheet”

    lol

    But I’m adding that to my list of howlingly stupid AI remarks, so thanks
  • ohnotnowohnotnow Posts: 3,775

    Leon said:

    Leon said:

    AI is just the next step of evolution of computing, it's not a revolution. It's the next tool, no more, no less.

    People have been using computers for decades. Computers have been automating elements for decades. AI is not some big unheard of thing, it's just what we already have but the next step along.

    AI is no more going to eliminate employment than computers or robots have done so.

    I agree with your assessment on the non-elimination of employment.

    Again, just on words, in the Academy, in computer science, we use the term "AI" to describe systems that we have been using for decades. We have AI currently: not the next step, but the present step. There is a journal called Artificial Intelligence, https://www.sciencedirect.com/journal/artificial-intelligence/issues , and that's been publishing since 1970.

    The way those AI systems work has changed, from logic-based systems to machine learning and now to generative AI. Their capabilities have increased. In the last few years, we've seen these leaps forward in a set of methods called generative AI, including generative AI systems for language, called large language models (LLMs). Any software that mimics some aspect of intelligence is called AI, even when it is a very limited intelligence.

    A lot of the discussion in this thread is using "AI" to mean something more, something akin to the term "artificial general intelligence" (AGI), i.e. something that can pass the Turing test. We don't have that now. I suspect we won't have that for awhile, although there have been a number of wild predictions suggesting otherwise.

    AI exists, is well-established and already in your smartphone. AI consciousness or AGI, that's another kettle of fish.


    You really think AI won’t eliminate lots of jobs? That’s brave

    In the last couple of months the bank Klarna started using OpenAI to replace customer service staff

    The result was so profound and “shocking” - their words - they put out a press statement as a kind of warning

    “New York, NY – February 27, 2024 – Klarna today announced its AI assistant powered by OpenAI. Now live globally for 1 month, the numbers speak for themselves:

    The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats

    It is doing the equivalent work of 700 full-time agents

    It is on par with human agents in regard to customer satisfaction score

    It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries

    Customers now resolve their errands in less than 2 mins compared to 11 mins previously

    It’s available in 23 markets, 24/7 and communicates in more than 35 languages

    It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024”

    That’s 700 jobs. In one go. In one modest company

    This is the beginning
    Which is bugger all.

    Its not one modest company, its a global company, and its the kind of crap jobs that are churned routinely, indeed its largely outsourced that kind of job anyway precisely because its so unimportant.
    ohnotnow said:

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.
    He’s not alone. Most of the people on this thread have no idea how quickly AI is progressing in all fields - hence, in part, their foolish opinions
    Or we understand it full well, but its moot.

    AI has been progressing for fifty years.

    Automation has been progressing for centuries.

    Automation, of which AI is another step of evolution, hasn't eliminated jobs net.

    Its eliminated some jobs, but then new ones are created, its evolution not revolution.

    AI is a tool, just like spreadsheets and the internet and a bazillion other things are tools. They're tools that will eliminate some menial jobs and create other ones, just like every other form of automation before it.
    It baffles me what a terrible job MS have done with their 'copilot for office'. It's really almost 'clippy' level bad. "Order this spreadsheet by column D" "Can't - soz". "Sum column Q" "Can't - soz". "Create a powerpoint about renewable energy" "Ok! Here's some pictures of a zebra and a few bullet-points about watermills!"
  • Leon said:

    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
    Yeah but so what?

    That kind of development has been happening for centuries.

    I get you want everything to be gamechanging, but life evolves. The job market evolves.

    Don't be a Luddite.
  • TrumanTruman Posts: 279
    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
  • MoonRabbitMoonRabbit Posts: 13,493

    Hester's comments were so bad they made me feel sorry for Diane Abbott.

    I really don’t understand this joke at all 🤷‍♀️ would you like to explain it to us?
  • Leon said:

    Leon said:

    Leon said:

    AI is just the next step of evolution of computing, it's not a revolution. It's the next tool, no more, no less.

    People have been using computers for decades. Computers have been automating elements for decades. AI is not some big unheard of thing, it's just what we already have but the next step along.

    AI is no more going to eliminate employment than computers or robots have done so.

    I agree with your assessment on the non-elimination of employment.

    Again, just on words, in the Academy, in computer science, we use the term "AI" to describe systems that we have been using for decades. We have AI currently: not the next step, but the present step. There is a journal called Artificial Intelligence, https://www.sciencedirect.com/journal/artificial-intelligence/issues , and that's been publishing since 1970.

    The way those AI systems work has changed, from logic-based systems to machine learning and now to generative AI. Their capabilities have increased. In the last few years, we've seen these leaps forward in a set of methods called generative AI, including generative AI systems for language, called large language models (LLMs). Any software that mimics some aspect of intelligence is called AI, even when it is a very limited intelligence.

    A lot of the discussion in this thread is using "AI" to mean something more, something akin to the term "artificial general intelligence" (AGI), i.e. something that can pass the Turing test. We don't have that now. I suspect we won't have that for awhile, although there have been a number of wild predictions suggesting otherwise.

    AI exists, is well-established and already in your smartphone. AI consciousness or AGI, that's another kettle of fish.


    You really think AI won’t eliminate lots of jobs? That’s brave

    In the last couple of months the bank Klarna started using OpenAI to replace customer service staff

    The result was so profound and “shocking” - their words - they put out a press statement as a kind of warning

    “New York, NY – February 27, 2024 – Klarna today announced its AI assistant powered by OpenAI. Now live globally for 1 month, the numbers speak for themselves:

    The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats

    It is doing the equivalent work of 700 full-time agents

    It is on par with human agents in regard to customer satisfaction score

    It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries

    Customers now resolve their errands in less than 2 mins compared to 11 mins previously

    It’s available in 23 markets, 24/7 and communicates in more than 35 languages

    It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024”

    That’s 700 jobs. In one go. In one modest company

    This is the beginning
    Which is bugger all.

    Its not one modest company, its a global company, and its the kind of crap jobs that are churned routinely, indeed its largely outsourced that kind of job anyway precisely because its so unimportant.
    ohnotnow said:

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.
    He’s not alone. Most of the people on this thread have no idea how quickly AI is progressing in all fields - hence, in part, their foolish opinions
    Or we understand it full well, but its moot.

    AI has been progressing for fifty years.

    Automation has been progressing for centuries.

    Automation, of which AI is another step of evolution, hasn't eliminated jobs net.

    Its eliminated some jobs, but then new ones are created, its evolution not revolution.

    AI is a tool, just like spreadsheets and the internet and a bazillion other things are tools. They're tools that will eliminate some menial jobs and create other ones, just like every other form of automation before it.
    AI is not “a tool just like a spreadsheet”

    lol

    But I’m adding that to my list of howlingly stupid AI remarks, so thanks
    AI absolutely is a tool.

    You want it to generate a bit of text, or a picture, you can do so. Tool.

    How you use that text, or picture, is up to you.
  • Scott_xPScott_xP Posts: 35,986
    @itvpeston
    “I would call an election, frankly, now… we are going backwards"

    Conservative peer and supermarket boss Lord Rose urges the PM to call a general election now, warning the UK is falling behind other countries and needs a leader with a mandate who can “get on with it”

    #Peston
  • Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    I didn't realise there was much mosque development in St Petersburg.

    The rest doesn't surprise me though.
  • ohnotnowohnotnow Posts: 3,775
    Taz said:

    On the subject of AI this is interestimg

    https://x.com/taylor_rosie/status/1767853702742774144?s=61

    I can only see the first tweet (X?) in the chain - can you give us the gist?
  • Northern_AlNorthern_Al Posts: 8,375
    Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    If they're privately racist, why are they telling you?
  • LostPasswordLostPassword Posts: 18,354
    edited March 13
    Is the racist/not-racist result with don't knows excluded, or are there really zero percent of people who don't know either way?
  • MoonRabbitMoonRabbit Posts: 13,493
    Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    How would you bet on UK election date, SovBot. May or Autumn?
  • JosiasJessopJosiasJessop Posts: 42,549
    Leon said:

    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
    You do not know what you're talking about. You are totally clueless. Worse, when people try to tell you where you're going wrong, you ignore them and continue riding the hype train down into the abyss.

    "where I can easily spend half a day catching up with AI developments"

    That's your problem. You read hype and believe it, and you don't have the intelligence to realise it's hype.

    You are a fool.
  • BenpointerBenpointer Posts: 34,653
    edited March 13
    kjh said:

    I have become a fruitarian. I only eat stuff that falls from trees. Yesterday I ate 3 apples and a squirrel.

    We were out with our dog in the woods a while back and a squirrel fell from a tree - hitting the ground right in front of us.

    I am not sure who was most surprised: us, the dog or the squirrel. The squirrel reacted quickest though and was back up the tree before the dog could pounce.
  • TazTaz Posts: 14,362
    edited March 13
    ohnotnow said:

    Taz said:

    On the subject of AI this is interestimg

    https://x.com/taylor_rosie/status/1767853702742774144?s=61

    I can only see the first tweet (X?) in the chain - can you give us the gist?
    Sure, a journalist put out a journo request for comment on a specific financial topic and got a response from a PR repressing three “people”:that were clearly chat GPT and she’s asking if others are experiencing it.

    The clients quoted seemed to be for crypto sites. But of the fake/scam variety. The PR, on checking, didn’t seem to exist. She concludes.

    “it seems like these (potential scam) sites are using AI to pick up journo requests, generate punchy comments and get quoted in mainstream media, making the sites appear legitimate. Judging by the "featured in" sections, others have fallen for it. V. worrying.”
  • BenpointerBenpointer Posts: 34,653

    Hester's comments were so bad they made me feel sorry for Diane Abbott.

    I really don’t understand this joke at all 🤷‍♀️ would you like to explain it to us?
    What makes you think it's a joke? It just seemed like a statement of fact to me.
  • MoonRabbitMoonRabbit Posts: 13,493

    Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    I didn't realise there was much mosque development in St Petersburg.

    The rest doesn't surprise me though.
    Why do we always presume they are from Russia, and not GBNews? 🤔
  • BenpointerBenpointer Posts: 34,653
    Sunak taking a racist's money is headline news on the BBC too.
  • Leon said:

    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
    You do not know what you're talking about. You are totally clueless. Worse, when people try to tell you where you're going wrong, you ignore them and continue riding the hype train down into the abyss.

    "where I can easily spend half a day catching up with AI developments"

    That's your problem. You read hype and believe it, and you don't have the intelligence to realise it's hype.

    You are a fool.
    We're supposed to think that masses of jobs are going to be eliminated due to us being able to type a prompt and get a picture out.

    But I've been able to do that with Google for about twenty years.

    Now the tool is generating the image rather than searching those it knows about, but its not exactly a revolution.

    The only people who thinks its a revolution is the mugs, and those who think they can profit off those mugs.
  • ohnotnowohnotnow Posts: 3,775

    Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    I didn't realise there was much mosque development in St Petersburg.

    The rest doesn't surprise me though.
    Why do we always presume they are from Russia, and not GBNews? 🤔
    Because they can usually spell?
  • Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    I didn't realise there was much mosque development in St Petersburg.

    The rest doesn't surprise me though.
    Why do we always presume they are from Russia, and not GBNews? 🤔
    What's the difference?
  • MalmesburyMalmesbury Posts: 50,212

    Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    I didn't realise there was much mosque development in St Petersburg.

    The rest doesn't surprise me though.


    Well, they do have this lovely fellow.
  • MoonRabbitMoonRabbit Posts: 13,493

    Hester's comments were so bad they made me feel sorry for Diane Abbott.

    I really don’t understand this joke at all 🤷‍♀️ would you like to explain it to us?
    What makes you think it's a joke? It just seemed like a statement of fact to me.
    Not a sarcastic comment on ‘National Be Nice to Di Day’ then?
  • LeonLeon Posts: 55,235

    Leon said:

    Leon said:

    Leon said:

    AI is just the next step of evolution of computing, it's not a revolution. It's the next tool, no more, no less.

    People have been using computers for decades. Computers have been automating elements for decades. AI is not some big unheard of thing, it's just what we already have but the next step along.

    AI is no more going to eliminate employment than computers or robots have done so.

    I agree with your assessment on the non-elimination of employment.

    Again, just on words, in the Academy, in computer science, we use the term "AI" to describe systems that we have been using for decades. We have AI currently: not the next step, but the present step. There is a journal called Artificial Intelligence, https://www.sciencedirect.com/journal/artificial-intelligence/issues , and that's been publishing since 1970.

    The way those AI systems work has changed, from logic-based systems to machine learning and now to generative AI. Their capabilities have increased. In the last few years, we've seen these leaps forward in a set of methods called generative AI, including generative AI systems for language, called large language models (LLMs). Any software that mimics some aspect of intelligence is called AI, even when it is a very limited intelligence.

    A lot of the discussion in this thread is using "AI" to mean something more, something akin to the term "artificial general intelligence" (AGI), i.e. something that can pass the Turing test. We don't have that now. I suspect we won't have that for awhile, although there have been a number of wild predictions suggesting otherwise.

    AI exists, is well-established and already in your smartphone. AI consciousness or AGI, that's another kettle of fish.


    You really think AI won’t eliminate lots of jobs? That’s brave

    In the last couple of months the bank Klarna started using OpenAI to replace customer service staff

    The result was so profound and “shocking” - their words - they put out a press statement as a kind of warning

    “New York, NY – February 27, 2024 – Klarna today announced its AI assistant powered by OpenAI. Now live globally for 1 month, the numbers speak for themselves:

    The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats

    It is doing the equivalent work of 700 full-time agents

    It is on par with human agents in regard to customer satisfaction score

    It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries

    Customers now resolve their errands in less than 2 mins compared to 11 mins previously

    It’s available in 23 markets, 24/7 and communicates in more than 35 languages

    It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024”

    That’s 700 jobs. In one go. In one modest company

    This is the beginning
    Which is bugger all.

    Its not one modest company, its a global company, and its the kind of crap jobs that are churned routinely, indeed its largely outsourced that kind of job anyway precisely because its so unimportant.
    ohnotnow said:

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.
    He’s not alone. Most of the people on this thread have no idea how quickly AI is progressing in all fields - hence, in part, their foolish opinions
    Or we understand it full well, but its moot.

    AI has been progressing for fifty years.

    Automation has been progressing for centuries.

    Automation, of which AI is another step of evolution, hasn't eliminated jobs net.

    Its eliminated some jobs, but then new ones are created, its evolution not revolution.

    AI is a tool, just like spreadsheets and the internet and a bazillion other things are tools. They're tools that will eliminate some menial jobs and create other ones, just like every other form of automation before it.
    AI is not “a tool just like a spreadsheet”

    lol

    But I’m adding that to my list of howlingly stupid AI remarks, so thanks
    AI absolutely is a tool.

    You want it to generate a bit of text, or a picture, you can do so. Tool.

    How you use that text, or picture, is up to you.
    No, I agree. It’s just a tool

    Like the Internet. As you say. Just a tool

    For me the internet is most like a vegetable spiralizer. They’re both tools that make things easier. One allows you instant access to all the accumulated knowledge of mankind and also lets you talk to any human being on earth, and the other one makes cucumbers whirly when you feed it in


    Exactly the same, basically

    From: “The Wit and Wisdom of Bartholomew Roberts, of Newent, Chapter one: How I somehow manage to tie my laces”
  • JosiasJessopJosiasJessop Posts: 42,549

    Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    I didn't realise there was much mosque development in St Petersburg.

    The rest doesn't surprise me though.
    "St Petersburg" was trending on Twitter the other day. I was surprised to find because the St Peterburg (Russia) had been attacked by drones, but those tweets were interspersed with others about the IndyCar race in St Petersburg (USA).
  • ydoethurydoethur Posts: 71,354
    I'm disappointed.I thought this was going to be an update about the Orange Haired Loon with the Small Mushroom Shaped Thingy was posting his latest half-billion dollar bond in New York.
  • MoonRabbitMoonRabbit Posts: 13,493

    Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    I didn't realise there was much mosque development in St Petersburg.

    The rest doesn't surprise me though.
    Why do we always presume they are from Russia, and not GBNews? 🤔
    What's the difference?
    Exactly!
  • ajbajb Posts: 147
    edited March 13
    Supposedly underwriters can be found to insure anything. I guess it must then be possible to insure against having to return a donation because the press find indefensible remarks by the donor. In the case of the conservatives, though, the premium might be too expensive.
  • ohnotnowohnotnow Posts: 3,775
    Taz said:

    ohnotnow said:

    Taz said:

    On the subject of AI this is interestimg

    https://x.com/taylor_rosie/status/1767853702742774144?s=61

    I can only see the first tweet (X?) in the chain - can you give us the gist?
    Sure, a journalist put out a journo request for comment on a specific financial topic and got a response from a PR repressing three “people”:that were clearly chat GPT and she’s asking if others are experiencing it.

    The clients quoted seemed to be for crypto sites. But of the fake/scam variety. The PR, on checking, didn’t seem to exist. She concludes.

    “it seems like these (potential scam) sites are using AI to pick up journo requests, generate punchy comments and get quoted in mainstream media, making the sites appear legitimate. Judging by the "featured in" sections, others have fallen for it. V. worrying.”
    Ah - gotcha. I've heard similar things from popular tech twitter accounts. Being bombarded with bot replies that sound somewhat plausible, but always interlinked into some sort of scam.

    On the other hand - they've also reported getting a lot of 'ChatGPT-feeling" emails from people who don't speak English well and wouldn't have felt confident reaching out without it's translation.

    Bit of a muddle in many ways.
  • Leon said:

    Leon said:

    Leon said:

    Leon said:

    AI is just the next step of evolution of computing, it's not a revolution. It's the next tool, no more, no less.

    People have been using computers for decades. Computers have been automating elements for decades. AI is not some big unheard of thing, it's just what we already have but the next step along.

    AI is no more going to eliminate employment than computers or robots have done so.

    I agree with your assessment on the non-elimination of employment.

    Again, just on words, in the Academy, in computer science, we use the term "AI" to describe systems that we have been using for decades. We have AI currently: not the next step, but the present step. There is a journal called Artificial Intelligence, https://www.sciencedirect.com/journal/artificial-intelligence/issues , and that's been publishing since 1970.

    The way those AI systems work has changed, from logic-based systems to machine learning and now to generative AI. Their capabilities have increased. In the last few years, we've seen these leaps forward in a set of methods called generative AI, including generative AI systems for language, called large language models (LLMs). Any software that mimics some aspect of intelligence is called AI, even when it is a very limited intelligence.

    A lot of the discussion in this thread is using "AI" to mean something more, something akin to the term "artificial general intelligence" (AGI), i.e. something that can pass the Turing test. We don't have that now. I suspect we won't have that for awhile, although there have been a number of wild predictions suggesting otherwise.

    AI exists, is well-established and already in your smartphone. AI consciousness or AGI, that's another kettle of fish.


    You really think AI won’t eliminate lots of jobs? That’s brave

    In the last couple of months the bank Klarna started using OpenAI to replace customer service staff

    The result was so profound and “shocking” - their words - they put out a press statement as a kind of warning

    “New York, NY – February 27, 2024 – Klarna today announced its AI assistant powered by OpenAI. Now live globally for 1 month, the numbers speak for themselves:

    The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats

    It is doing the equivalent work of 700 full-time agents

    It is on par with human agents in regard to customer satisfaction score

    It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries

    Customers now resolve their errands in less than 2 mins compared to 11 mins previously

    It’s available in 23 markets, 24/7 and communicates in more than 35 languages

    It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024”

    That’s 700 jobs. In one go. In one modest company

    This is the beginning
    Which is bugger all.

    Its not one modest company, its a global company, and its the kind of crap jobs that are churned routinely, indeed its largely outsourced that kind of job anyway precisely because its so unimportant.
    ohnotnow said:

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.
    He’s not alone. Most of the people on this thread have no idea how quickly AI is progressing in all fields - hence, in part, their foolish opinions
    Or we understand it full well, but its moot.

    AI has been progressing for fifty years.

    Automation has been progressing for centuries.

    Automation, of which AI is another step of evolution, hasn't eliminated jobs net.

    Its eliminated some jobs, but then new ones are created, its evolution not revolution.

    AI is a tool, just like spreadsheets and the internet and a bazillion other things are tools. They're tools that will eliminate some menial jobs and create other ones, just like every other form of automation before it.
    AI is not “a tool just like a spreadsheet”

    lol

    But I’m adding that to my list of howlingly stupid AI remarks, so thanks
    AI absolutely is a tool.

    You want it to generate a bit of text, or a picture, you can do so. Tool.

    How you use that text, or picture, is up to you.
    No, I agree. It’s just a tool

    Like the Internet. As you say. Just a tool

    For me the internet is most like a vegetable spiralizer. They’re both tools that make things easier. One allows you instant access to all the accumulated knowledge of mankind and also lets you talk to any human being on earth, and the other one makes cucumbers whirly when you feed it in


    Exactly the same, basically

    From: “The Wit and Wisdom of Bartholomew Roberts, of Newent, Chapter one: How I somehow manage to tie my laces”
    Yes, they're both tools. They both have their purpose.

    Want to know where you've seen an actress before - check IMDB, your spiraliser isn't much good for that.

    Want to cook some courgetti - use your spiraliser, going online isn't much good for that.

    Humanity has invented tons of tools over the millenia. Some more useful, some less.

    AI is just continued evolution along those lines.

    Its not like the advent of the internet has led us to having 50% unemployment currently, is it?
  • LostPasswordLostPassword Posts: 18,354

    Is the racist/not-racist result with don't knows excluded, or are there really zero percent of people who don't know either way?

    Okay. Found the data tables. Looks like the remark was presented to poll respondents who were asked to tick all the statements that they thought applied. So the 32% are the people who didn't tick the "The remark was racist" box, so will include people who aren't sure whether it is racist.

    Only 13% ticked the box "The remark had nothing to do with her gender nor colour of skin".
  • TheScreamingEaglesTheScreamingEagles Posts: 119,620
    ydoethur said:

    I'm disappointed.I thought this was going to be an update about the Orange Haired Loon with the Small Mushroom Shaped Thingy was posting his latest half-billion dollar bond in New York.

    I don't think I will write any more Trump headers as I don't think I will beat 'Ayrshire hotelier wins the GOP nomination.'
  • TimSTimS Posts: 12,984
    The survey results in the header are interesting. 32% don’t think Frank Hester’s remarks were racist, but only 18% think the Tories don’t have a problem with racism.

    So on average poll respondents think the Tories are more racist than Frank Hester.
  • geoffwgeoffw Posts: 8,708
    The AI revolution is a disruptor. Like the Industrial Revolution two centuries ago. Lots of jobs/occupations disappear. But others appear, some new some old. People will shift into activities where they have a comparative advantage. Imagine that AI bots can do everything humans can, but better or more efficiently. Even so, there will - with logical certainty - be activites where humans have comparative advantage, as David Ricardo taught us before the Industrial Revolution (though the Luddites of the time were oblivious to this insight).
  • LostPasswordLostPassword Posts: 18,354
    edited March 13
    TimS said:

    The survey results in the header are interesting. 32% don’t think Frank Hester’s remarks were racist, but only 18% think the Tories don’t have a problem with racism.

    So on average poll respondents think the Tories are more racist than Frank Hester.

    See my post at 9pm.

    Edit: And find the days tables yourself, here: https://jlpartners.com/polling-results
  • kle4kle4 Posts: 96,071
    edited March 13

    It was a story on PM today. Curious to why journalists aren't asking Hester why he gave £10 million to the Conservatives?

    Well that's no mystery - to buy influence and probably try to get a peerage at some point, there's no point in asking more.

    Yet another reason I would make donating more than £1000 to a political party prevent you from being considered for a peerage until two parliaments (or 8 years, whichever is longer) has passed. You aren't prevented from giving more money if you want, it just ensures there is no possible appearance of having bought a peerage.
  • kle4kle4 Posts: 96,071

    DavidL said:

    As I said yesterday, the real problem here is who is going to make a significant donation to any political party and risk being Hestered on the basis of something you allegedly said years before? Even trying to buy a seat in the House of Lords is surely not worth that.

    All political parties will feel the draught of this.

    Good. What a lot of decent money pissed up walls on a load of old shyster consultants.
    If we introduce measures (like preventing the 'reward' of donors) and that leads to parties not having as much cash, what harm? Do they need need large amounts of cash to get by? Maybe they should try to slim things down or get more members.
  • ydoethurydoethur Posts: 71,354

    ydoethur said:

    I'm disappointed.I thought this was going to be an update about the Orange Haired Loon with the Small Mushroom Shaped Thingy was posting his latest half-billion dollar bond in New York.

    I don't think I will write any more Trump headers as I don't think I will beat 'Ayrshire hotelier wins the GOP nomination.'
    Let us hope the tide will turn - berry soon.
  • LeonLeon Posts: 55,235

    Leon said:

    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
    You do not know what you're talking about. You are totally clueless. Worse, when people try to tell you where you're going wrong, you ignore them and continue riding the hype train down into the abyss.

    "where I can easily spend half a day catching up with AI developments"

    That's your problem. You read hype and believe it, and you don't have the intelligence to realise it's hype.

    You are a fool.
    Do you agree with @BartholomewRoberts that the Internet is exactly the same as a vegetable spiraliser?

  • ydoethur said:

    I'm disappointed.I thought this was going to be an update about the Orange Haired Loon with the Small Mushroom Shaped Thingy was posting his latest half-billion dollar bond in New York.

    I don't think I will write any more Trump headers as I don't think I will beat 'Ayrshire hotelier wins the GOP nomination.'
    'Ayrshire hotelier loses presidential election' would top it.
  • Scott_xPScott_xP Posts: 35,986
    @mikeysmith

    Rory Bremner tries out a new Rishi Sunak impression gag at the Best for Britain annual dinner: “What’s the difference between my dishwasher and my wife? My wife’s properly loaded.”
  • ydoethurydoethur Posts: 71,354

    ydoethur said:

    I'm disappointed.I thought this was going to be an update about the Orange Haired Loon with the Small Mushroom Shaped Thingy was posting his latest half-billion dollar bond in New York.

    I don't think I will write any more Trump headers as I don't think I will beat 'Ayrshire hotelier wins the GOP nomination.'
    'Ayrshire hotelier loses presidential election' would top it.
    A turn-ip for the ages.
  • FF43FF43 Posts: 17,208
    Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    Looked at the tables here to see if young people were more likely to think the remarks not racist https://home.38degrees.org.uk/2024/03/13/conservative-party-has-a-racism-problem-say-most-brits-as-voters-tell-party-to-reject-frank-hester-donation/

    There's a serious methodology problem with this poll in my view. Interviewees were asked to "tick all that apply" including that the remarks were racist. An unticked box is counted as a No in the tables above. But it's at least as likely to be no opinion/not interested.
  • FF43 said:

    Truman said:

    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    You would be surprised by how many young people are privately racist now. And its not like the old boomer racism its much harder edged than that.One guy i spoke to said he would likely start to carry a gun in 10 years to protect himself. There is much anger about the rampant building of mosques too.
    Looked at the tables here to see if young people were more likely to think the remarks not racist https://home.38degrees.org.uk/2024/03/13/conservative-party-has-a-racism-problem-say-most-brits-as-voters-tell-party-to-reject-frank-hester-donation/

    There's a serious methodology problem with this poll in my view. Interviewees were asked to "tick all that apply" including that the remarks were racist. An unticked box is counted as a No in the tables above. But it's at least as likely to be no opinion/not interested.
    That is awful methodology.
  • FrankBoothFrankBooth Posts: 9,826
    Professor of Medicine Martin Kurldorff says he has been fired from Harvard for telling the truth.

    https://www.city-journal.org/article/harvard-tramples-the-truth
  • LeonLeon Posts: 55,235
    geoffw said:

    The AI revolution is a disruptor. Like the Industrial Revolution two centuries ago. Lots of jobs/occupations disappear. But others appear, some new some old. People will shift into activities where they have a comparative advantage. Imagine that AI bots can do everything humans can, but better or more efficiently. Even so, there will - with logical certainty - be activites where humans have comparative advantage, as David Ricardo taught us before the Industrial Revolution (though the Luddites of the time were oblivious to this insight).

    Why is it a logical certainty that humans will be able to do jobs AI can’t?

    I’m not on a wind up. Properly curious

    Also what are these jobs? Lots of people would like to know including my daughters in their late teens now choosing uni courses

    I am fairly sure some jobs will survive and some will be created - but u don’t see it as “logically certain” - and I am much less sure that the new jobs will be anywhere near as plentiful as all the jobs lost
  • SeaShantyIrish2SeaShantyIrish2 Posts: 17,559
    OT - Could it be, that certain percentage of respondents, answered that Conservatives do NOT have a problem with Tory MP racism, because they do NOT believe that racism itself is a problem?

    Note that historically, many racists have believed - and many still do - that racism is a POSITIVE value, and NOT problematic . . . EXCEPT for potential backlash.

  • LostPasswordLostPassword Posts: 18,354

    OT - Could it be, that certain percentage of respondents, answered that Conservatives do NOT have a problem with Tory MP racism, because they do NOT believe that racism itself is a problem?

    Note that historically, many racists have believed - and many still do - that racism is a POSITIVE value, and NOT problematic . . . EXCEPT for potential backlash.

    You could do with having five options:
    Racist and that's a bad thing
    Racist and that's a good thing
    Not-racist and that's a good thing
    Not racist and that's a bad thing
    Don't know
  • Luckyguy1983Luckyguy1983 Posts: 28,411

    OT - Could it be, that certain percentage of respondents, answered that Conservatives do NOT have a problem with Tory MP racism, because they do NOT believe that racism itself is a problem?

    Note that historically, many racists have believed - and many still do - that racism is a POSITIVE value, and NOT problematic . . . EXCEPT for potential backlash.

    Might they then think the Tories have a racism problem because they're not racist enough?
  • FrankBoothFrankBooth Posts: 9,826
    Should we be taking this poll seriously. Does it meet the usual standards of the BPC?
  • rottenboroughrottenborough Posts: 62,693
    Julian Knight MP
    @julianknight15
    ·
    3h
    The Conservative Party must return every penny of Mr Hester’s donation no ifs no buts.

    https://twitter.com/julianknight15/status/1767977096645595317
  • mwadamsmwadams Posts: 3,588

    Leon said:

    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
    You do not know what you're talking about. You are totally clueless. Worse, when people try to tell you where you're going wrong, you ignore them and continue riding the hype train down into the abyss.

    "where I can easily spend half a day catching up with AI developments"

    That's your problem. You read hype and believe it, and you don't have the intelligence to realise it's hype.

    You are a fool.
    Whereas if you actually use this stuff you discover the glacial pace of productionisation, and how limited in scope the most impressive advances actually are.

    One of the most amazing pieces of "AI" tech is Excel autofill which involves building and tuning a custom model on the fly - the researchers involved are probably heading for a medal. But it isn't flashy and it isn't called 'Trevor' and it actually works 99% of the time (it's one of those <5min human supervised tasks I was talking about).

    But it is genuinely interesting and novel AI tech that has a huge amount of potential for further development.
  • SeaShantyIrish2SeaShantyIrish2 Posts: 17,559

    OT - Could it be, that certain percentage of respondents, answered that Conservatives do NOT have a problem with Tory MP racism, because they do NOT believe that racism itself is a problem?

    Note that historically, many racists have believed - and many still do - that racism is a POSITIVE value, and NOT problematic . . . EXCEPT for potential backlash.

    Might they then think the Tories have a racism problem because they're not racist enough?
    That is also a logical possibility.
  • Leon said:

    geoffw said:

    The AI revolution is a disruptor. Like the Industrial Revolution two centuries ago. Lots of jobs/occupations disappear. But others appear, some new some old. People will shift into activities where they have a comparative advantage. Imagine that AI bots can do everything humans can, but better or more efficiently. Even so, there will - with logical certainty - be activites where humans have comparative advantage, as David Ricardo taught us before the Industrial Revolution (though the Luddites of the time were oblivious to this insight).

    Why is it a logical certainty that humans will be able to do jobs AI can’t?

    I’m not on a wind up. Properly curious

    Also what are these jobs? Lots of people would like to know including my daughters in their late teens now choosing uni courses

    I am fairly sure some jobs will survive and some will be created - but u don’t see it as “logically certain” - and I am much less sure that the new jobs will be anywhere near as plentiful as all the jobs lost
    Because if you put your mind to it, there are clearly jobs that suit humans well.

    Which is why centuries of automation has pruned boring, monotonous, repetitive or dangerous jobs while instead boosting the availability of human-facing jobs.

    And that's not changing with AI.

    We will continue in the evolution of getting rid of jobs that don't need doing, and create new ones to replace them that now do instead, just as we've done for centuries. Just as happened with the rise of the internet, the rise of computers, the rise of robotics and everything else.
  • SeaShantyIrish2SeaShantyIrish2 Posts: 17,559

    Leon said:

    Leon said:

    Leon said:

    AI is just the next step of evolution of computing, it's not a revolution. It's the next tool, no more, no less.

    People have been using computers for decades. Computers have been automating elements for decades. AI is not some big unheard of thing, it's just what we already have but the next step along.

    AI is no more going to eliminate employment than computers or robots have done so.

    I agree with your assessment on the non-elimination of employment.

    Again, just on words, in the Academy, in computer science, we use the term "AI" to describe systems that we have been using for decades. We have AI currently: not the next step, but the present step. There is a journal called Artificial Intelligence, https://www.sciencedirect.com/journal/artificial-intelligence/issues , and that's been publishing since 1970.

    The way those AI systems work has changed, from logic-based systems to machine learning and now to generative AI. Their capabilities have increased. In the last few years, we've seen these leaps forward in a set of methods called generative AI, including generative AI systems for language, called large language models (LLMs). Any software that mimics some aspect of intelligence is called AI, even when it is a very limited intelligence.

    A lot of the discussion in this thread is using "AI" to mean something more, something akin to the term "artificial general intelligence" (AGI), i.e. something that can pass the Turing test. We don't have that now. I suspect we won't have that for awhile, although there have been a number of wild predictions suggesting otherwise.

    AI exists, is well-established and already in your smartphone. AI consciousness or AGI, that's another kettle of fish.


    You really think AI won’t eliminate lots of jobs? That’s brave

    In the last couple of months the bank Klarna started using OpenAI to replace customer service staff

    The result was so profound and “shocking” - their words - they put out a press statement as a kind of warning

    “New York, NY – February 27, 2024 – Klarna today announced its AI assistant powered by OpenAI. Now live globally for 1 month, the numbers speak for themselves:

    The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats

    It is doing the equivalent work of 700 full-time agents

    It is on par with human agents in regard to customer satisfaction score

    It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries

    Customers now resolve their errands in less than 2 mins compared to 11 mins previously

    It’s available in 23 markets, 24/7 and communicates in more than 35 languages

    It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024”

    That’s 700 jobs. In one go. In one modest company

    This is the beginning
    Which is bugger all.

    Its not one modest company, its a global company, and its the kind of crap jobs that are churned routinely, indeed its largely outsourced that kind of job anyway precisely because its so unimportant.
    ohnotnow said:

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.
    He’s not alone. Most of the people on this thread have no idea how quickly AI is progressing in all fields - hence, in part, their foolish opinions
    Or we understand it full well, but its moot.

    AI has been progressing for fifty years.

    Automation has been progressing for centuries.

    Automation, of which AI is another step of evolution, hasn't eliminated jobs net.

    Its eliminated some jobs, but then new ones are created, its evolution not revolution.

    AI is a tool, just like spreadsheets and the internet and a bazillion other things are tools. They're tools that will eliminate some menial jobs and create other ones, just like every other form of automation before it.
    AI is not “a tool just like a spreadsheet”

    lol

    But I’m adding that to my list of howlingly stupid AI remarks, so thanks
    AI absolutely is a tool.

    You want it to generate a bit of text, or a picture, you can do so. Tool.

    How you use that text, or picture, is up to you.
    As far as I can tell, one of the major applications of AI at present, is jazzing up crap YouTube vids.
  • Leon said:

    Leon said:

    Leon said:

    AI is just the next step of evolution of computing, it's not a revolution. It's the next tool, no more, no less.

    People have been using computers for decades. Computers have been automating elements for decades. AI is not some big unheard of thing, it's just what we already have but the next step along.

    AI is no more going to eliminate employment than computers or robots have done so.

    I agree with your assessment on the non-elimination of employment.

    Again, just on words, in the Academy, in computer science, we use the term "AI" to describe systems that we have been using for decades. We have AI currently: not the next step, but the present step. There is a journal called Artificial Intelligence, https://www.sciencedirect.com/journal/artificial-intelligence/issues , and that's been publishing since 1970.

    The way those AI systems work has changed, from logic-based systems to machine learning and now to generative AI. Their capabilities have increased. In the last few years, we've seen these leaps forward in a set of methods called generative AI, including generative AI systems for language, called large language models (LLMs). Any software that mimics some aspect of intelligence is called AI, even when it is a very limited intelligence.

    A lot of the discussion in this thread is using "AI" to mean something more, something akin to the term "artificial general intelligence" (AGI), i.e. something that can pass the Turing test. We don't have that now. I suspect we won't have that for awhile, although there have been a number of wild predictions suggesting otherwise.

    AI exists, is well-established and already in your smartphone. AI consciousness or AGI, that's another kettle of fish.


    You really think AI won’t eliminate lots of jobs? That’s brave

    In the last couple of months the bank Klarna started using OpenAI to replace customer service staff

    The result was so profound and “shocking” - their words - they put out a press statement as a kind of warning

    “New York, NY – February 27, 2024 – Klarna today announced its AI assistant powered by OpenAI. Now live globally for 1 month, the numbers speak for themselves:

    The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats

    It is doing the equivalent work of 700 full-time agents

    It is on par with human agents in regard to customer satisfaction score

    It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries

    Customers now resolve their errands in less than 2 mins compared to 11 mins previously

    It’s available in 23 markets, 24/7 and communicates in more than 35 languages

    It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024”

    That’s 700 jobs. In one go. In one modest company

    This is the beginning
    Which is bugger all.

    Its not one modest company, its a global company, and its the kind of crap jobs that are churned routinely, indeed its largely outsourced that kind of job anyway precisely because its so unimportant.
    ohnotnow said:

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.
    He’s not alone. Most of the people on this thread have no idea how quickly AI is progressing in all fields - hence, in part, their foolish opinions
    Or we understand it full well, but its moot.

    AI has been progressing for fifty years.

    Automation has been progressing for centuries.

    Automation, of which AI is another step of evolution, hasn't eliminated jobs net.

    Its eliminated some jobs, but then new ones are created, its evolution not revolution.

    AI is a tool, just like spreadsheets and the internet and a bazillion other things are tools. They're tools that will eliminate some menial jobs and create other ones, just like every other form of automation before it.
    AI is not “a tool just like a spreadsheet”

    lol

    But I’m adding that to my list of howlingly stupid AI remarks, so thanks
    AI absolutely is a tool.

    You want it to generate a bit of text, or a picture, you can do so. Tool.

    How you use that text, or picture, is up to you.
    As far as I can tell, one of the major applications of AI at present, is jazzing up crap YouTube vids.
    Yes, its good at churning out BS.

    As a purveyor of BS, you can see why Leon is concerned he might lose his job.

    Most people have more than BS as their output at work though.
  • JosiasJessopJosiasJessop Posts: 42,549
    edited March 13
    mwadams said:

    Leon said:

    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
    You do not know what you're talking about. You are totally clueless. Worse, when people try to tell you where you're going wrong, you ignore them and continue riding the hype train down into the abyss.

    "where I can easily spend half a day catching up with AI developments"

    That's your problem. You read hype and believe it, and you don't have the intelligence to realise it's hype.

    You are a fool.
    Whereas if you actually use this stuff you discover the glacial pace of productionisation, and how limited in scope the most impressive advances actually are.

    One of the most amazing pieces of "AI" tech is Excel autofill which involves building and tuning a custom model on the fly - the researchers involved are probably heading for a medal. But it isn't flashy and it isn't called 'Trevor' and it actually works 99% of the time (it's one of those 5min human supervised tasks I was talking about).

    But it is genuinely interesting and novel AI tech that has a huge amount of potential for further development.
    Oh, there will certainly be many uses for this tech - and it's good tech. But it's nowhere near as good as the bedwetters make out, or as revolutionary.

    I'm bearish about it; my own view is that the current LLMs are an evolutionary dead-end towards AGI or anything much better. I might be wrong about that, but I'd need some convincing.

    Edit, interestingly, the less than symbol before the '5min' in your post was stopping a reply from posting, perhaps because Vanilla thought it was the start of a tag.
  • mwadamsmwadams Posts: 3,588

    Leon said:

    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
    You do not know what you're talking about. You are totally clueless. Worse, when people try to tell you where you're going wrong, you ignore them and continue riding the hype train down into the abyss.

    "where I can easily spend half a day catching up with AI developments"

    That's your problem. You read hype and believe it, and you don't have the intelligence to realise it's hype.

    You are a fool.

    I don't think Leon is a fool, though. I think he is doing his research on the internet and is listening to the "most interesting" voices, rather than the "most sober" voices. The sober voices are probably too sober. But in this space (as in crypto before) the most interesting voices are, unfortunately, mostly grifters and the odd fruit loop.
  • mwadamsmwadams Posts: 3,588

    mwadams said:

    Leon said:

    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
    You do not know what you're talking about. You are totally clueless. Worse, when people try to tell you where you're going wrong, you ignore them and continue riding the hype train down into the abyss.

    "where I can easily spend half a day catching up with AI developments"

    That's your problem. You read hype and believe it, and you don't have the intelligence to realise it's hype.

    You are a fool.
    Whereas if you actually use this stuff you discover the glacial pace of productionisation, and how limited in scope the most impressive advances actually are.

    One of the most amazing pieces of "AI" tech is Excel autofill which involves building and tuning a custom model on the fly - the researchers involved are probably heading for a medal. But it isn't flashy and it isn't called 'Trevor' and it actually works 99% of the time (it's one of those 5min human supervised tasks I was talking about).

    But it is genuinely interesting and novel AI tech that has a huge amount of potential for further development.
    Oh, there will certainly be many uses for this tech - and it's good tech. But it's nowhere near as good as the bedwetters make out, or as revolutionary.

    I'm bearish about it; my own view is that the current LLMs are an evolutionary dead-end towards AGI or anything much better. I might be wrong about that, but I'd need some convincing.
    Oh, I agree that they are absolutely a dead end for AGI. I'm not even thinking about that. But LLMs (especially if we can tame their energy demands with SLMs and 1.58bit models) are going to be ever more useful. When narrowed in scope from pretending to have Genuine People Personalities.
  • nico679nico679 Posts: 6,275
    Some of the public might think it’s not realistic to return this donation . It’s a shame we didn’t have a question on whether the Tories should accept future donations from Hester .
  • JosiasJessopJosiasJessop Posts: 42,549
    mwadams said:

    Leon said:

    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
    You do not know what you're talking about. You are totally clueless. Worse, when people try to tell you where you're going wrong, you ignore them and continue riding the hype train down into the abyss.

    "where I can easily spend half a day catching up with AI developments"

    That's your problem. You read hype and believe it, and you don't have the intelligence to realise it's hype.

    You are a fool.

    I don't think Leon is a fool, though. I think he is doing his research on the internet and is listening to the "most interesting" voices, rather than the "most sober" voices. The sober voices are probably too sober. But in this space (as in crypto before) the most interesting voices are, unfortunately, mostly grifters and the odd fruit loop.
    IMV listening to only the 'most interesting' voices' rather than the 'most sober' ones as well is a good indication of a fool! for the reasons you put in your last sentence....
  • turbotubbsturbotubbs Posts: 17,392

    Professor of Medicine Martin Kurldorff says he has been fired from Harvard for telling the truth.

    https://www.city-journal.org/article/harvard-tramples-the-truth

    Scanned some of the article and it’s clear he is a lockdown skeptic, which is fine. But he rails against closing schools as the covid risks were min8mal to children. Well yes, they were. But what about all the older folks, you know the teachers, support staff etc? I do have to wonder at some peoples logical abilities.
  • Luckyguy1983Luckyguy1983 Posts: 28,411

    Leon said:

    Leon said:

    Leon said:

    AI is just the next step of evolution of computing, it's not a revolution. It's the next tool, no more, no less.

    People have been using computers for decades. Computers have been automating elements for decades. AI is not some big unheard of thing, it's just what we already have but the next step along.

    AI is no more going to eliminate employment than computers or robots have done so.

    I agree with your assessment on the non-elimination of employment.

    Again, just on words, in the Academy, in computer science, we use the term "AI" to describe systems that we have been using for decades. We have AI currently: not the next step, but the present step. There is a journal called Artificial Intelligence, https://www.sciencedirect.com/journal/artificial-intelligence/issues , and that's been publishing since 1970.

    The way those AI systems work has changed, from logic-based systems to machine learning and now to generative AI. Their capabilities have increased. In the last few years, we've seen these leaps forward in a set of methods called generative AI, including generative AI systems for language, called large language models (LLMs). Any software that mimics some aspect of intelligence is called AI, even when it is a very limited intelligence.

    A lot of the discussion in this thread is using "AI" to mean something more, something akin to the term "artificial general intelligence" (AGI), i.e. something that can pass the Turing test. We don't have that now. I suspect we won't have that for awhile, although there have been a number of wild predictions suggesting otherwise.

    AI exists, is well-established and already in your smartphone. AI consciousness or AGI, that's another kettle of fish.


    You really think AI won’t eliminate lots of jobs? That’s brave

    In the last couple of months the bank Klarna started using OpenAI to replace customer service staff

    The result was so profound and “shocking” - their words - they put out a press statement as a kind of warning

    “New York, NY – February 27, 2024 – Klarna today announced its AI assistant powered by OpenAI. Now live globally for 1 month, the numbers speak for themselves:

    The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats

    It is doing the equivalent work of 700 full-time agents

    It is on par with human agents in regard to customer satisfaction score

    It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries

    Customers now resolve their errands in less than 2 mins compared to 11 mins previously

    It’s available in 23 markets, 24/7 and communicates in more than 35 languages

    It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024”

    That’s 700 jobs. In one go. In one modest company

    This is the beginning
    Which is bugger all.

    Its not one modest company, its a global company, and its the kind of crap jobs that are churned routinely, indeed its largely outsourced that kind of job anyway precisely because its so unimportant.
    ohnotnow said:

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.
    He’s not alone. Most of the people on this thread have no idea how quickly AI is progressing in all fields - hence, in part, their foolish opinions
    Or we understand it full well, but its moot.

    AI has been progressing for fifty years.

    Automation has been progressing for centuries.

    Automation, of which AI is another step of evolution, hasn't eliminated jobs net.

    Its eliminated some jobs, but then new ones are created, its evolution not revolution.

    AI is a tool, just like spreadsheets and the internet and a bazillion other things are tools. They're tools that will eliminate some menial jobs and create other ones, just like every other form of automation before it.
    AI is not “a tool just like a spreadsheet”

    lol

    But I’m adding that to my list of howlingly stupid AI remarks, so thanks
    AI absolutely is a tool.

    You want it to generate a bit of text, or a picture, you can do so. Tool.

    How you use that text, or picture, is up to you.
    As far as I can tell, one of the major applications of AI at present, is jazzing up crap YouTube vids.
    Yes, its good at churning out BS.

    As a purveyor of BS, you can see why Leon is concerned he might lose his job.

    Most people have more than BS as their output at work though.
    Leon writes BS quite well though, and as far as my attempts with ai are concerned, it writes average to poor copy, and takes quite a lot of prompting to reach average.
  • FrankBoothFrankBooth Posts: 9,826

    Professor of Medicine Martin Kurldorff says he has been fired from Harvard for telling the truth.

    https://www.city-journal.org/article/harvard-tramples-the-truth

    Scanned some of the article and it’s clear he is a lockdown skeptic, which is fine. But he rails against closing schools as the covid risks were min8mal to children. Well yes, they were. But what about all the older folks, you know the teachers, support staff etc? I do have to wonder at some peoples logical abilities.
    You can disagree with his argument. The point is he has been fired.
  • RochdalePioneersRochdalePioneers Posts: 28,895
    “Of course it’s not racist to say Diane Abbott makes him want to shoot all black women. He didn’t use the N word, did he? So what’s all the fuss about?

    Next on GB News, why Muslim monsters should be lynched”
  • kinabalukinabalu Posts: 42,098
    FF43 said:

    32% think Frank Hester's remark wasn't racist.

    What would he have to say for them to think, yeah that really is racist?

    They would need something such as "I think black people are inferior".

    Of course people who do think that never say it. On a self-reporting basis racism is a thing of the past.
  • LeonLeon Posts: 55,235
    I’ve made it to the death place of Bolivar. And I’m having one of those evil cold light fizzy lagers again

    And I’m afraid it is fucking delicious, after 6 hours on a Colombian bus




    Not a bad view as well


  • SeaShantyIrish2SeaShantyIrish2 Posts: 17,559

    Leon said:

    Leon said:

    Leon said:

    AI is just the next step of evolution of computing, it's not a revolution. It's the next tool, no more, no less.

    People have been using computers for decades. Computers have been automating elements for decades. AI is not some big unheard of thing, it's just what we already have but the next step along.

    AI is no more going to eliminate employment than computers or robots have done so.

    I agree with your assessment on the non-elimination of employment.

    Again, just on words, in the Academy, in computer science, we use the term "AI" to describe systems that we have been using for decades. We have AI currently: not the next step, but the present step. There is a journal called Artificial Intelligence, https://www.sciencedirect.com/journal/artificial-intelligence/issues , and that's been publishing since 1970.

    The way those AI systems work has changed, from logic-based systems to machine learning and now to generative AI. Their capabilities have increased. In the last few years, we've seen these leaps forward in a set of methods called generative AI, including generative AI systems for language, called large language models (LLMs). Any software that mimics some aspect of intelligence is called AI, even when it is a very limited intelligence.

    A lot of the discussion in this thread is using "AI" to mean something more, something akin to the term "artificial general intelligence" (AGI), i.e. something that can pass the Turing test. We don't have that now. I suspect we won't have that for awhile, although there have been a number of wild predictions suggesting otherwise.

    AI exists, is well-established and already in your smartphone. AI consciousness or AGI, that's another kettle of fish.


    You really think AI won’t eliminate lots of jobs? That’s brave

    In the last couple of months the bank Klarna started using OpenAI to replace customer service staff

    The result was so profound and “shocking” - their words - they put out a press statement as a kind of warning

    “New York, NY – February 27, 2024 – Klarna today announced its AI assistant powered by OpenAI. Now live globally for 1 month, the numbers speak for themselves:

    The AI assistant has had 2.3 million conversations, two-thirds of Klarna’s customer service chats

    It is doing the equivalent work of 700 full-time agents

    It is on par with human agents in regard to customer satisfaction score

    It is more accurate in errand resolution, leading to a 25% drop in repeat inquiries

    Customers now resolve their errands in less than 2 mins compared to 11 mins previously

    It’s available in 23 markets, 24/7 and communicates in more than 35 languages

    It’s estimated to drive a $40 million USD in profit improvement to Klarna in 2024”

    That’s 700 jobs. In one go. In one modest company

    This is the beginning
    Which is bugger all.

    Its not one modest company, its a global company, and its the kind of crap jobs that are churned routinely, indeed its largely outsourced that kind of job anyway precisely because its so unimportant.
    ohnotnow said:

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.
    He’s not alone. Most of the people on this thread have no idea how quickly AI is progressing in all fields - hence, in part, their foolish opinions
    Or we understand it full well, but its moot.

    AI has been progressing for fifty years.

    Automation has been progressing for centuries.

    Automation, of which AI is another step of evolution, hasn't eliminated jobs net.

    Its eliminated some jobs, but then new ones are created, its evolution not revolution.

    AI is a tool, just like spreadsheets and the internet and a bazillion other things are tools. They're tools that will eliminate some menial jobs and create other ones, just like every other form of automation before it.
    AI is not “a tool just like a spreadsheet”

    lol

    But I’m adding that to my list of howlingly stupid AI remarks, so thanks
    AI absolutely is a tool.

    You want it to generate a bit of text, or a picture, you can do so. Tool.

    How you use that text, or picture, is up to you.
    As far as I can tell, one of the major applications of AI at present, is jazzing up crap YouTube vids.
    Yes, its good at churning out BS.

    As a purveyor of BS, you can see why Leon is concerned he might lose his job.

    Most people have more than BS as their output at work though.
    Nevertheless, demand for BS is insatiable, and has been throughout human history. And no doubt pre-history.

    For example, strongly suspect that prophet Ezekiel's "great wheels of fire" as recorded in Old Testament, may well have been early tech inventor-investor hype about an allegedly revolutionary breakthrough in wheeled-vehicle technology.

    Which of course was NOT achieved for several millennia afterward. Biblical inerrancy notwithstanding . . . or rather slightly delayed . . .
  • Scott_xPScott_xP Posts: 35,986
    @mikeysmith
    Rory Bremner: “I try not to hate Tory donors, but I look at Frank Hester and he makes me hate all Tory donors. He should be shot…”
  • geoffwgeoffw Posts: 8,708
    Leon said:

    geoffw said:

    The AI revolution is a disruptor. Like the Industrial Revolution two centuries ago. Lots of jobs/occupations disappear. But others appear, some new some old. People will shift into activities where they have a comparative advantage. Imagine that AI bots can do everything humans can, but better or more efficiently. Even so, there will - with logical certainty - be activites where humans have comparative advantage, as David Ricardo taught us before the Industrial Revolution (though the Luddites of the time were oblivious to this insight).

    Why is it a logical certainty that humans will be able to do jobs AI can’t?

    I’m not on a wind up. Properly curious

    Also what are these jobs? Lots of people would like to know including my daughters in their late teens now choosing uni courses

    I am fairly sure some jobs will survive and some will be created - but u don’t see it as “logically certain” - and I am much less sure that the new jobs will be anywhere near as plentiful as all the jobs lost
    The AI machines may be able to do everything better than any human. They therefore have an absolute productivity advantage in every job. But humans, though less productive at any activity than the bots, are bound to be relatively better at some than others. It is a matter of relative (or in Ricardian language, comparative) costs across different activities (jobs). The usual explicator for this argument is in the field of international trade, comparing Portugal with England, where Portugal can produce both wine and cloth more efficiently than England. Nevertheless both gain from trade by producing the product in which they have camparative advantage.
    Wiki's explanation with Ricardo's example is here: https://en.wikipedia.org/wiki/Comparative_advantage
  • RochdalePioneersRochdalePioneers Posts: 28,895

    I accept it's not particularly scientific but on the whole, if Leon is raving.about.something it's safe to conclude it's a dud.

    its.a.dud
  • SeaShantyIrish2SeaShantyIrish2 Posts: 17,559

    OT - Could it be, that certain percentage of respondents, answered that Conservatives do NOT have a problem with Tory MP racism, because they do NOT believe that racism itself is a problem?

    Note that historically, many racists have believed - and many still do - that racism is a POSITIVE value, and NOT problematic . . . EXCEPT for potential backlash.

    You could do with having five options:
    Racist and that's a bad thing
    Racist and that's a good thing
    Not-racist and that's a good thing
    Not racist and that's a bad thing
    Don't know
    Option 6: "Get out of my face, and go back where you came from!"
  • rkrkrkrkrkrk Posts: 8,293
    Leon said:

    geoffw said:

    The AI revolution is a disruptor. Like the Industrial Revolution two centuries ago. Lots of jobs/occupations disappear. But others appear, some new some old. People will shift into activities where they have a comparative advantage. Imagine that AI bots can do everything humans can, but better or more efficiently. Even so, there will - with logical certainty - be activites where humans have comparative advantage, as David Ricardo taught us before the Industrial Revolution (though the Luddites of the time were oblivious to this insight).

    Why is it a logical certainty that humans will be able to do jobs AI can’t?

    I’m not on a wind up. Properly curious

    Also what are these jobs? Lots of people would like to know including my daughters in their late teens now choosing uni courses

    I am fairly sure some jobs will survive and some will be created - but u don’t see it as “logically certain” - and I am much less sure that the new jobs will be anywhere near as plentiful as all the jobs lost
    I struggle to imagine the terminally ill devout Catholic requesting an AI robot to play a recorded message for their last rites.

    Perhaps not that helpful for your daughter though!
  • JohnLilburneJohnLilburne Posts: 6,240
    Leon said:

    I’ve made it to the death place of Bolivar. And I’m having one of those evil cold light fizzy lagers again

    And I’m afraid it is fucking delicious, after 6 hours on a Colombian bus




    Not a bad view as well


    I always assumed you travelled a bit more lux. My Thai and Cambodian bus trips weren't too bad though. Bangkok to Kanchanaburi on 3rd class non-air-con train was a blast
  • LeonLeon Posts: 55,235
    mwadams said:

    Leon said:

    ohnotnow said:

    (Apols about a FPT, but I posted about 30 seconds after the new thread message as per..)

    Ratters said:

    AI hype is a classic case of engineers/coders/tech bros thinking that what most people do in their job on a day basis can just be automated by their latest invention.

    We were supposed to have fully automated self driving cars ago. They seem to be no closer to solving "the last 1%" than they were then, even in more friendly locations in the US vs somewhere harder like London.

    Assisted driving makes life easier for long drives on the motorway. But it's not the revolution of being able to be a passenger idly watching TV that was promised.

    These LLMs will similarly help in creating first drafts of essays, of emails, of music etc. It will make things much more efficient. Some people will lose their jobs. But it won't be a wholesale revolution making 25% of the population redundant.

    Finally - I don't think AGI necessarily means enlightened genius that takes over everything. The current Tory cabinet has general intelligence, but that doesn't make it any less crap.

    And I don't care what 'experts in the field say' as they have made so many incorrect predictions on other forms of AI in the recent past. Only the passage of time will prove me or them to be right or wrong.

    *Ducks*

    Yup.

    The “AI” we have at the moment is fair at some things. It is hopeless at detailed accuracy.

    What it can do is spot patterns or the absence of patterns. So give it a zillion bank transactions and it can spot anomalies. Not all of them, but for fraud detection it’s quite useful for creating alerts for humans to check out.

    An interesting one is giving an LLM a human written essay and asking it what insights and the themes within the topic were *left out*
    It's not just detailed accuracy, it's *any* accuracy. Hence hands with six or seven fingers, or five Beatles, or athletes with missing forearms.

    Indeed, current 'AI' has no idea of the concept of accuracy. Or fingers. Or anything.
    I'm terribly sorry - but that's incorrect. Unless your idea of 'AI' is typing something into ChatGPT/Dalle and then eye-rolling that it wasn't what you wanted.

    The field has progressed well beyond that.

    To be fair to @JosiasJessop he’s not alone. Most of the commenters on the prior thread have no idea of how quickly AI advances in multiple sectors. This explains - in part - the many foolish opinions

    Indeed I have some sympathy. I have the kind of leisured job where I can easily spend half a day catching up with AI developments (and I regularly do). Also I am paid for this; as I am then paid for my opinions on AI

    But if you’re a normal person without many hours to spare wandering the world and reading reddit then AI is advancing so rapidly you can’t keep up. We are getting revelatory improvements almost daily now

    This is what it is like to be on an exponential curve
    You do not know what you're talking about. You are totally clueless. Worse, when people try to tell you where you're going wrong, you ignore them and continue riding the hype train down into the abyss.

    "where I can easily spend half a day catching up with AI developments"

    That's your problem. You read hype and believe it, and you don't have the intelligence to realise it's hype.

    You are a fool.

    I don't think Leon is a fool, though. I think he is doing his research on the internet and is listening to the "most interesting" voices, rather than the "most sober" voices. The sober voices are probably too sober. But in this space (as in crypto before) the most interesting voices are, unfortunately, mostly grifters and the odd fruit loop.
    I read the skeptics as well as the hype-mongers. I read Gary Marcus as well as Kurzweill and Hinton

    The thing I’ve noticed is that the skeptics have been consistently wrong and sometimes humiliatingly so, for the last couple of years. Which says quite a lot

    Eg Yann Lecun. Goes onstage in Dubai and says text-to-video is impossible as things stand, it’s far harder than text-to-image, it needs a new architecture (which he is designing!) blah blah

    Two days later OpenAI announce Sora and make him look like a total fool

    I also think tech people and engineers are sometimes the most clueless about what they themselves are designing. It’s not in their intellectual remit

    Would you ask a guy building one of the first carburetors what will be the impact of the internal combustion engine on American urbanism?

    Of course not. He won’t have a scooby. You need someone with a vastly wider frame of reference. A Renaissance man who has travelled widely and slept with literally hundreds of young women
  • turbotubbsturbotubbs Posts: 17,392

    Professor of Medicine Martin Kurldorff says he has been fired from Harvard for telling the truth.

    https://www.city-journal.org/article/harvard-tramples-the-truth

    Scanned some of the article and it’s clear he is a lockdown skeptic, which is fine. But he rails against closing schools as the covid risks were min8mal to children. Well yes, they were. But what about all the older folks, you know the teachers, support staff etc? I do have to wonder at some peoples logical abilities.
    You can disagree with his argument. The point is he has been fired.
    I don’t know why he was fired but there will be reasons. Generally if you do things that put your institution in a bad light there are consequences. Does he say why he has been fired (other than for speaking the’truth’)?
  • LeonLeon Posts: 55,235

    Leon said:

    I’ve made it to the death place of Bolivar. And I’m having one of those evil cold light fizzy lagers again

    And I’m afraid it is fucking delicious, after 6 hours on a Colombian bus




    Not a bad view as well


    I always assumed you travelled a bit more lux. My Thai and Cambodian bus trips weren't too bad though. Bangkok to Kanchanaburi on 3rd class non-air-con train was a blast
    Bus is the only way around Colombia. Don’t think they even have trains

    Also I hate - seriously - travelling lux all the time. How boring. I like to switch it up. Bus and train stations are great places to meet people and check out local life. Even the lowlife. And if you rough it for a few days then you appreciate the luxury (like these suites) all the more
  • TrumanTruman Posts: 279

    OT - Could it be, that certain percentage of respondents, answered that Conservatives do NOT have a problem with Tory MP racism, because they do NOT believe that racism itself is a problem?

    Note that historically, many racists have believed - and many still do - that racism is a POSITIVE value, and NOT problematic . . . EXCEPT for potential backlash.

    Might they then think the Tories have a racism problem because they're not racist enough?
    That is also a logical possibility.
    The tories certainly arent racist enough for some of their voters. Granted we are talking a small pool of people now.
  • TrumanTruman Posts: 279
    Very disappointing attendance at Cheltenham races. Sign of the times.

    Official attendance at Cheltenham today was 46,771.

    Comparing to previous Wednesdays, it’s down from 50,387 in 2023 and 64,431 in 2022 (first year after pandemic).

    Worrying.

    #Cheltenham | #CheltenhamFestival
    Last edited
    7:26 PM · Mar 13, 2024
    ·
    134.8K
    Views
  • LeonLeon Posts: 55,235
    rkrkrk said:

    Leon said:

    geoffw said:

    The AI revolution is a disruptor. Like the Industrial Revolution two centuries ago. Lots of jobs/occupations disappear. But others appear, some new some old. People will shift into activities where they have a comparative advantage. Imagine that AI bots can do everything humans can, but better or more efficiently. Even so, there will - with logical certainty - be activites where humans have comparative advantage, as David Ricardo taught us before the Industrial Revolution (though the Luddites of the time were oblivious to this insight).

    Why is it a logical certainty that humans will be able to do jobs AI can’t?

    I’m not on a wind up. Properly curious

    Also what are these jobs? Lots of people would like to know including my daughters in their late teens now choosing uni courses

    I am fairly sure some jobs will survive and some will be created - but u don’t see it as “logically certain” - and I am much less sure that the new jobs will be anywhere near as plentiful as all the jobs lost
    I struggle to imagine the terminally ill devout Catholic requesting an AI robot to play a recorded message for their last rites.

    Perhaps not that helpful for your daughter though!
    Vicar or hooker

    Those are the two jobs that will be left
  • rottenboroughrottenborough Posts: 62,693

    “Of course it’s not racist to say Diane Abbott makes him want to shoot all black women. He didn’t use the N word, did he? So what’s all the fuss about?

    Next on GB News, why Muslim monsters should be lynched”

    This weekend's polling will be interesting.

    Tories below 20%?

    Dire opening few minutes for Sunak on BBC news at 10.
  • TrumanTruman Posts: 279
    And this. People are starting to hurt.

    That’s the official attendance but being there today I’d confidently say it was much lower.
    I’ve never seen so it quiet in the last 12 years I’ve been going.
    The ridiculous ticket, alcohol and food prices are extortionate and people are finally waking up to it
    9:21 PM · Mar 13, 2024
    ·
    343
    Views

    https://x.com/LeeWoodhouse00/status/1768024518071640470?s=20
  • LeonLeon Posts: 55,235
    geoffw said:

    Leon said:

    geoffw said:

    The AI revolution is a disruptor. Like the Industrial Revolution two centuries ago. Lots of jobs/occupations disappear. But others appear, some new some old. People will shift into activities where they have a comparative advantage. Imagine that AI bots can do everything humans can, but better or more efficiently. Even so, there will - with logical certainty - be activites where humans have comparative advantage, as David Ricardo taught us before the Industrial Revolution (though the Luddites of the time were oblivious to this insight).

    Why is it a logical certainty that humans will be able to do jobs AI can’t?

    I’m not on a wind up. Properly curious

    Also what are these jobs? Lots of people would like to know including my daughters in their late teens now choosing uni courses

    I am fairly sure some jobs will survive and some will be created - but u don’t see it as “logically certain” - and I am much less sure that the new jobs will be anywhere near as plentiful as all the jobs lost
    The AI machines may be able to do everything better than any human. They therefore have an absolute productivity advantage in every job. But humans, though less productive at any activity than the bots, are bound to be relatively better at some than others. It is a matter of relative (or in Ricardian language, comparative) costs across different activities (jobs). The usual explicator for this argument is in the field of international trade, comparing Portugal with England, where Portugal can produce both wine and cloth more efficiently than England. Nevertheless both gain from trade by producing the product in which they have camparative advantage.
    Wiki's explanation with Ricardo's example is here: https://en.wikipedia.org/wiki/Comparative_advantage
    I remember that from my economics A level. I’m not at all sure the rules of international trade can be applied to the advent of AI, in fact I’m fairly sure they can’t
This discussion has been closed.