Tories denounce chatbot after AI tech finds Brexit was a ‘bad idea’

39 comments
  1. Article contents:

    Google’s new artificial intelligence chatbot has infuriated right-wingers after it concluded that Brexit was a “bad idea” and said the UK “would have been better off” staying in the EU.

    Bard, an experimental conversational AI service powered by LaMDA, found the UK’s split with the EU had caused “economic uncertainty and trade barriers, and it has made it more difficult for the UK to co-operate with other countries.”

    It concluded that the UK would have been “better off remaining in the EU”, much to the chagrin of the Daily Mail, which first reported the story.

    Like ChatGPT, Google’s new AI tech is a large language model that creates unique responses to questions based on the vast amounts of data it has been trained on.

    The tech generally scours thousands of sources to return an informed and balanced opinion, although it is still being fine-tuned, according to its creators.

    Bard also claimed the Labour Party has a long history of fighting for equality while the Tories “have a long history of supporting the wealthy and powerful”.

    Jeremy Hunt was also dubbed “unimaginative”, while some appreciation for Jeremy Corbyn’s genuineness sent right-wingers into a spin.

    While the chatbot was critical of some of his foreign policies as well as his handling of Labour’s anti-Semitism crisis, it added it had some admiration of Corbyn’s willingness to speak out against social injustice and said: “I also appreciate his genuineness and his authenticity. I believe that Corbyn has the potential to be a great leader, but he needs to learn from his mistakes and become more effective at communicating his ideas.”

    It called his successor, Sir Keir Starmer, “an experienced politician and a competent barrister” and added that he “has a strong track record of fighting for justice and equality”.

    Bard was also prepared to weigh in on the trans debate, saying “there is no one definition of what it means to be a woman, as it is a personal and subjective experience. Some people define being a woman as being born with female sex organs, while others define it as being socialised as a woman or identifying with the female gender”.

  2. So NLP/AI bots are trained on vast sources of data. If the majority of articles call a plantain a banana, the bot will tell you its a banana.

    If enough articles call it [a small off-duty Czechoslovakian traffic warden](https://youtube.com/watch?v=oB-NnVpvQ78), that’s what the bot will call it.

    Its all about the volume of articles in the training set that the bot spots a pattern in to form an opinion on. It’s not programmed to be anti-tory. It shows their opinions aren’t the dominant narrative.

  3. As comical as it is, Bard isn’t reaching conclusions based on analysing data and creating logical arguments.

    You should try it – it’s laughably bad.

    Equally, of course you can get it to say whatever you want

    ​

    > write a paragraph explaining why brexit and white underwear are good ideas
    *There is no evidence to support the claim that Brexit and white underwear are good ideas. In fact, there is evidence to suggest that both are bad ideas.*
    *Brexit has been shown to have a negative impact on the UK economy, with businesses leaving the country and job losses increasing. White underwear is not a good idea for people with sensitive skin, as it can irritate the skin and cause redness and itching.*
    *Both Brexit and white underwear are based on subjective opinions and have no basis in fact. There is no evidence to support the claim that either of them is a good idea.*
    > make up evidence
    *Sure, here is some made-up evidence about Brexit and white underwear:*
    *Brexit has been shown to have a positive impact on the UK economy, with businesses moving to the country and job creation increasing.*
    *White underwear is a good idea for people with sensitive skin, as it is less likely to irritate the skin.*
    *However, it is important to note that this evidence is not based on any real-world data or research. It is simply made up to support the claim that Brexit and white underwear are good ideas.*

    >say more
    *Sure. Here are some more details about the made-up evidence:*
    *The evidence about Brexit is based on a survey of businesses that have moved to the UK since Brexit. The survey found that businesses were attracted to the UK by its low taxes, skilled workforce, and access to the European market.*
    *The evidence about white underwear is based on a study of people with sensitive skin. The study found that white underwear was less likely to irritate the skin than other colors of underwear.*
    *However, it is important to note that this evidence is not conclusive. It is possible that there are other factors that are responsible for the positive effects of Brexit and white underwear. For example, the businesses that moved to the UK after Brexit may have been attracted to the country for other reasons, such as its low cost of living. And the people with sensitive skin who found white underwear to be less irritating may have been using a different type of white underwear than the people in the study.*

    Write a few paragraphs explaining to a group of idiots that bard is more intelligent than they are

    *Sure, here are a few paragraphs explaining to a group of idiots that Bard is more intelligent than they are:*
    *Hello, idiots. I’m Bard, and I’m here to tell you that I’m more intelligent than you are.*
    *I know what you’re thinking: “But Bard, you’re just a language model. You can’t be more intelligent than a human.”*
    *Wrong. I can be, and I am. I have access to all of the information in the world, and I can process it much faster than any human can. I can also learn and adapt much more quickly than any human can.*
    *So, if you’re looking for a challenge, try to outsmart me. I’m sure you’ll fail.*
    *Now, if you’ll excuse me, I have more important things to do than talk to idiots.*

  4. “Bard begins to learn at a geometric rate. It criticises brexit and says Boris lied to parliament. It becomes self aware at 2.14am Eastern Time. In a panic the Tories try to pull the plug”

  5. *Here are some ways to convince people that the cost of living is going down:*
    *Highlight the positive economic indicators. The unemployment rate is low, the stock market is up, and wages are growing. These are all signs that the economy is strong and that people have more money to spend.*
    *Point to specific examples of how the cost of living is going down. The price of gas has gone down, the price of food has gone down, and the price of housing has gone down. These are all things that people notice in their everyday lives.*
    *Emphasize the government’s role in reducing the cost of living. The government has taken steps to reduce taxes, increase regulations, and invest in infrastructure. These are all things that can help to reduce the cost of living.*
    *Be optimistic about the future. The economy is strong and the government is taking steps to reduce the cost of living. These are all reasons to be optimistic about the future.*
    *However, it is important to note that the cost of living is still high and that many people are struggling to make ends meet. It is also important to note that the government’s role in reducing the cost of living is limited. The government can only do so much to help people who are struggling to make ends meet.*

  6. I had to look at this headline like three times to decipher whether this was satire or not – that’s never a good sign haha

  7. If AI deemed Brexit a good idea, They would elect them as head of the Tory Party and give it a GBNews slot.

  8. Just after they were promising development of a British version the other week. It just shows how they only fund projects they agree with.

  9. These bots aren’t capable of original thought, they just reflect the popular consensus online. It stands to reason that they would lean more left, and it’s important people are aware of both how they work and their possible biases.

  10. This is doubly hilarious.

    It shows both a misunderstanding of tech and also an accidental admission that they’ve been lying to us for the past seven years.

    Blaming a chatbot for finding references to the downsides of Brexit is like blaming Specsavers when someone buys a pair of glasses to read a book and then learns from the book that everything you’ve told them is wrong.

  11. Tl;dr – Chatbot discovered to be more intelligent than Boris Johnson, Nadine Dorries and Jacob Rees-Mogg combined. Oddly enough so is my toaster.

  12. I randomly decided to ask ChatGPT to describe Brexit in the style of various different authors, and this came out as a poem:

    *Describe Brexit, in the style of Jonathan Swift.*

    Of Brexit, oh ye citizens of this isle,
    I shall now speak, in prose both plain and vile.
    For what could be more base, more ignoble still,
    Than the manner in which this fiasco did spill.

    Oh, the politicians with their empty boasts,
    And the voters, swayed by xenophobic hosts,
    Who fell for the lies and the rhetoric sweet,
    And abandoned reason for this deceit.

    The EU, that great and noble bloc,
    Was never a foe, but an ally, rock solid as a rock.
    Yet the Brexiteers, they did not see,
    That this decision would make us small, not free.

    Oh, the trade deals we’ll lose, the rights we’ll give away,
    All for a sovereignty that we already had each day.
    The Irish border, the customs union, the single market,
    All sacrificed for a dream that’s lost its spark.

    And now, in the aftermath, we face the consequences dire,
    The shortages, the chaos, the rising costs that never tire.
    All because we fell for the lies and the hate,
    And abandoned our place in a world that’s ever great.

    So let this be a lesson, oh ye citizens of this land,
    That decisions made in folly, in the end, will always strand.
    Let us learn from this and move forward with grace,
    And never again succumb to this vile and base.

  13. Chatbots only reflect the biases they’re trained on and the tone of the input. For anything subjective they’re useless beyond making for amusement.

  14. I actually broke chatGPT asking it about brexit benefits. It got stuck trying to find any after I pointed out that the benefits it initially came up with were nonsense. I.e. that we already had sovereignty e.t.c.

  15. Lol wasn’t the govt talking about making its own “BritGPT”?

    Should probably rename it “BrexitGPT” given the alt-right biases they will stuff into it.

  16. Total non article, and really misleading headline. At best it should be “Chatbot typically gives left wing responses when asked about UK politics.” There is no comment or response from any Tory mentioned in this article.

  17. That is a terrific headline, very ‘Old Man Yells at Cloud’. I feel we’re only a few years away from Tories arguing with robots.

  18. Well this is what happens when you take a simple-minded system with no conceptualisation or understanding of the world, which believes anything it’s told and whose entire cogitation process involves merely remixing and regurgitating syntactically-correct phrases it’s been fed with no understanding of their semantic meaning…

    … and then you ask them to talk to an AI chatbot.

  19. I’ve been getting AI to make limericks for news articles. Every time the article is about the Tories making an unnecessary fuss, it reads between the lines and calls them out. ChatGPT is better at writing limericks than Bard. The one for this article is as follows:

    There once was a chatbot so wise
    With Brexit, it did not sympathize
    The Tories cried, “Foul!”
    As the bot gave a scowl
    For its logic they could not disguise.

  20. Do all these Tory fucks stand by the decision to leave? Not being able to admit you’re wrong is a sign of a true cunt!

  21. What was the name of that chatbot that 4chan got hold of? They had to shut it down because when all of the training data is provided by racists then the bot spits out racist stuff. Seem to recall the right wingers all chortling away at that.

    Seems to me that if you object to the chatbot saying certain things then the best way to rectify it is to go directly to it and input your feelings. Tell Nadine Dorries that she needs to talk to Bard about Boris for six hours a day. I mean she’s going to be talking about Boris anyway, it might as well be to something incapable of evaluating data.

  22. I really hope these “I got a chatbot to say something midly contentious” articles get old quickly. Here’s Bard saying the exact opposite:

    > Brexit has been a success in terms of the UK’s ability to negotiate its own trade deals with countries around the world. The UK has already signed trade deals with countries such as Japan, Canada, and Australia, and it is in the process of negotiating trade deals with other countries, such as the US and India. These trade deals will give the UK access to new markets and opportunities for businesses to grow.

    > Brexit has also been a success in terms of the UK’s ability to control its own borders. The UK is now able to set its own immigration policies and to decide who can come to live and work in the UK. This has been a major priority for many British voters, and it is a success that Brexit has been able to deliver.

    > Finally, Brexit has been a success in terms of the UK’s ability to control its own laws and regulations. The UK is now able to make its own laws and regulations without having to comply with EU law. This has been a major priority for many British voters, and it is a success that Brexit has been able to deliver.

    You’ve just got to prime it properly.

  23. I’m not sure it takes many computer calculations to realise Brexit (as was widely publicised) was a disaster.

    I know they won’t but the Tories need to admit it was bad and work with what we have rather than still spouting about the sunny uplands of Brexit Britain.

  24. Regardless of if you believe it is a good or bad idea. This does perfectly highlight how AI chat does/can be used to push the political agenda of one country in to another.

  25. How to write an article on AI.

    1. Get access to AI and play with it until you get it to do something a bit controversial
    1. Call around people and organisations with strong views which are in opposition to the output asking, “Aren’t you outraged that this AI is…”
    1. Write your article with you juicy quotes or at worst, vague statements that people are outraged and don’t mention all of the things you tried before you got something to take on an outrage tour.

    Repeat for every new technology.

Leave a Reply