Bravo, Sam. This will change how I think with and use AI tools for the better. Iād go as far as recommending this as Required reading for any AI users!
Oh wow, thanks so much, Billy. This means a great deal to me. I have to admit that this paper really took me back as well, and as a result, I've been questioning my own AI practises on almost a daily basis.
Interesting as always. Claude still makes enough mistakes that I do not have any perceived inferiority. Just today it thought 6.5% was a low point # when there was 4.3% in the mix as well. And I'll keep š out for the others. Don't incept me Claude!
Thanks JHong! Do you know, I think this is really effective practise, looking out for those silly mistakes that our AI tools make every single day, just to remind us that humans should always be in the lead and also that the limits of our own capacity and creativity have no bounds.
Thank you, Sam, for flagging this. Scary stuff, indeed.
As a work/org design person, I am also looking at what Kroyer & LĆ¼tge call second-order amplifiers, and then mainly (1) technology and (3) situation/organisation (I find the "individual amplifiers" section the weakest part of the paper). After just cursory reading my take/advice would be (1) de-gamify, and (3) de-pressurize. In other words: Slow AI.
Thank you Dr Troxler! The de-gamify is very interesting to me, because I think that gamification is a great misjustice that companies use to take advantage of our playful natures. I think the literature calls this 'playbour', which is a term I very much like.
Phew, Iām so glad I use AI as little as possible. I find it appalling, especially in certain contexts where itās used intensively, such as in the genocide in Palestine. I think thatās the only way theyāve been able to kill so many children ā by eliminating all human capacity for empathy and compassion. I suppose the soldiers who do this can sleep at night precisely because the responsibility is no longer theirs and it becomes like a video game.
Thank you, Irene. Even though this is beyond my particular field of expertise, I do think that there needs to be a serious conversation around AI, agency, and warfare and what that means for the use of autonomous vehicles in any armed conflict. In my honest opinion, it requires a rewrite of the Geneva Convention, but it's perhaps for other experts to comment on.
Itās not mine either. But you donāt need to be an expert or a philosopher specialising in ethics for your own body to feel a deep sense of revulsion. In his book *Humankind*, Rutger Bregman recounts how many soldiers in the Second World War were unable to fire their weapons. It seems that in these modern wars, the aim is to avoid this by stripping away the human layer that prevents inhuman decisions from being made.
Bravo, Sam. This will change how I think with and use AI tools for the better. Iād go as far as recommending this as Required reading for any AI users!
Oh wow, thanks so much, Billy. This means a great deal to me. I have to admit that this paper really took me back as well, and as a result, I've been questioning my own AI practises on almost a daily basis.
Interesting as always. Claude still makes enough mistakes that I do not have any perceived inferiority. Just today it thought 6.5% was a low point # when there was 4.3% in the mix as well. And I'll keep š out for the others. Don't incept me Claude!
Thanks JHong! Do you know, I think this is really effective practise, looking out for those silly mistakes that our AI tools make every single day, just to remind us that humans should always be in the lead and also that the limits of our own capacity and creativity have no bounds.
Thank you, Sam, for flagging this. Scary stuff, indeed.
As a work/org design person, I am also looking at what Kroyer & LĆ¼tge call second-order amplifiers, and then mainly (1) technology and (3) situation/organisation (I find the "individual amplifiers" section the weakest part of the paper). After just cursory reading my take/advice would be (1) de-gamify, and (3) de-pressurize. In other words: Slow AI.
Maybe worth a couple of additional posts.
Thank you Dr Troxler! The de-gamify is very interesting to me, because I think that gamification is a great misjustice that companies use to take advantage of our playful natures. I think the literature calls this 'playbour', which is a term I very much like.
Phew, Iām so glad I use AI as little as possible. I find it appalling, especially in certain contexts where itās used intensively, such as in the genocide in Palestine. I think thatās the only way theyāve been able to kill so many children ā by eliminating all human capacity for empathy and compassion. I suppose the soldiers who do this can sleep at night precisely because the responsibility is no longer theirs and it becomes like a video game.
Thank you, Irene. Even though this is beyond my particular field of expertise, I do think that there needs to be a serious conversation around AI, agency, and warfare and what that means for the use of autonomous vehicles in any armed conflict. In my honest opinion, it requires a rewrite of the Geneva Convention, but it's perhaps for other experts to comment on.
Itās not mine either. But you donāt need to be an expert or a philosopher specialising in ethics for your own body to feel a deep sense of revulsion. In his book *Humankind*, Rutger Bregman recounts how many soldiers in the Second World War were unable to fire their weapons. It seems that in these modern wars, the aim is to avoid this by stripping away the human layer that prevents inhuman decisions from being made.