3 ms·

Would love a short explanation/link to your MA thesis, that seems very interesting!

syntaxfree 5y ago
Hubert Dreyfus worked on Heideggerean critiques of AI while at MIT rubbing shoulders with Marvin Minsky and the like. There’s a good documentary called “Being in the world”; last time I checked it had been posted on YouTube.
Mark_Kumleben 5y ago
For sure! Of course, the official website has it paywalled, so here's a Google Docs link: https://docs.google.com/document/d/1WL1F-sEazLeo8uFjNpPyFLaktinB3vhAxCtJ74vLYWI/edit?usp=sharing https://docs.google.com/document/d/1WL1F-sEazLeo8uFjNpPyFLak... The Heidegger stuff is in the fourth section, 'Artificial Intelligence In Technological Society', but a rough summary: Heidegger argues that the essence of technology involves a command to human beings to order the world as standing-reserves of resources. Our specific place in this schema is as the being who can do the work of ordering, of revealing the world as orderable standing-reserve and then actually putting it into a particular order (e.g. of a supply chain). However, modern AI, with its emphasis on pattern-recognition, search, optimizing for a goal, etc., is also working towards that function. This has two implications. First, it suggests why AI research heads in this direction, because it's the way human intelligence manifests in the technological framework (and, thus, suggests that AI research could potentially benefit from trying to replicate other aspects of the human mind). Secondly, it means that modern AI might affect a Heideggerian theory of technology by making it possible for humans to delegate the work of ordering to AI, which wouldn't have appeared a possibility in Heidegger's day. If we can in fact hand off the task of ordering to AI, how should we do that and what effects might it have on human culture? While it appears a loss of agency, it may even have a liberating effect in the same way other technologies have liberated us from previous forms of labour.
- pmoriarty 5y ago
  But Heidegger is against seeing the world merely as a standing-reserve of resources. That's a utilitarian, reductionist, objectifying view that's the antithesis of the view that Heidegger advocates.
  - zelias 5y ago
    I disagree. You can separate the author's observations (namely, the identified pattern of revealing standing reserve) from the author's intentions (standing reserve = bad). However, one can easily conceive of an AI that "obsoletes" humanity from the supply chain, so-to-speak, which would theoretically allow humans unbridled opportunity to "be", as Heidegger envisions it. Once human labor is obsolete, what will become of humanity?
    - pmoriarty 5y ago
      Will play also be obsolete? Because work can be play, and play can be work. Work often has a negative connotation in today's world, but it can also be positive, fun, and rewarding. The economic whip (ie. the incentive of working to survive) might become obsolete, but that doesn't mean that people will stop wanting to work for fun, fulfillment, or to make original contributions. It's unlikely many people would willingly choose to be garbagemen and toilet cleaners, and we'll probably all breathe a sigh of relief when those jobs are finally automated, but will we stop wanting to make art or music when AI can make it "just as good"? Even in more practical fields than art or music, it's more likely that we'll see AI-human collaborations rather than AI doing everything by itself without human input. What is human is constantly in flux anyway. Even if we consider just technology's effect on the human, there are humans now with all sorts of prosthesis that make them "more than human" in some ways, and this trend itself will likely continue, with things like more and more advanced neural interfaces, memory upgrades, sensory enhancement, etc. The future of humanity will likely be as some sort of cyborg.
      - IntrepidWorm 5y ago
        I think back to Calhoun's rat utopia experiments and wonder if humans could ever be happy in a post-work society. https://www.smithsonianmag.com/smart-news/how-mouse-utopias-1960s-led-grim-predictions-humans-180954423/ https://www.smithsonianmag.com/smart-news/how-mouse-utopias-... While the predictions made by the study regarding human behavior have since been called into question (and because a human version of these experiments would be horribly unethical), it does bear to think about what the world would look like if all work was performed by automated means free of human interaction. On one hand, there are many jobs that would be seldom missed by most people. On the other, people can find fullfillment in just about anything.

Hubert Dreyfus worked on Heideggerean critiques of AI while at MIT rubbing shoulders with Marvin Minsky and the like. There’s a good documentary called “Being in the world”; last time I checked it had been posted on YouTube.

Mark_Kumleben 5y ago

For sure! Of course, the official website has it paywalled, so here's a Google Docs link: https://docs.google.com/document/d/1WL1F-sEazLeo8uFjNpPyFLaktinB3vhAxCtJ74vLYWI/edit?usp=sharing https://docs.google.com/document/d/1WL1F-sEazLeo8uFjNpPyFLak... The Heidegger stuff is in the fourth section, 'Artificial Intelligence In Technological Society', but a rough summary: Heidegger argues that the essence of technology involves a command to human beings to order the world as standing-reserves of resources. Our specific place in this schema is as the being who can do the work of ordering, of revealing the world as orderable standing-reserve and then actually putting it into a particular order (e.g. of a supply chain). However, modern AI, with its emphasis on pattern-recognition, search, optimizing for a goal, etc., is also working towards that function. This has two implications. First, it suggests why AI research heads in this direction, because it's the way human intelligence manifests in the technological framework (and, thus, suggests that AI research could potentially benefit from trying to replicate other aspects of the human mind). Secondly, it means that modern AI might affect a Heideggerian theory of technology by making it possible for humans to delegate the work of ordering to AI, which wouldn't have appeared a possibility in Heidegger's day. If we can in fact hand off the task of ordering to AI, how should we do that and what effects might it have on human culture? While it appears a loss of agency, it may even have a liberating effect in the same way other technologies have liberated us from previous forms of labour.

pmoriarty 5y ago
But Heidegger is against seeing the world merely as a standing-reserve of resources. That's a utilitarian, reductionist, objectifying view that's the antithesis of the view that Heidegger advocates.
- zelias 5y ago
  I disagree. You can separate the author's observations (namely, the identified pattern of revealing standing reserve) from the author's intentions (standing reserve = bad). However, one can easily conceive of an AI that "obsoletes" humanity from the supply chain, so-to-speak, which would theoretically allow humans unbridled opportunity to "be", as Heidegger envisions it. Once human labor is obsolete, what will become of humanity?
  - pmoriarty 5y ago
    Will play also be obsolete? Because work can be play, and play can be work. Work often has a negative connotation in today's world, but it can also be positive, fun, and rewarding. The economic whip (ie. the incentive of working to survive) might become obsolete, but that doesn't mean that people will stop wanting to work for fun, fulfillment, or to make original contributions. It's unlikely many people would willingly choose to be garbagemen and toilet cleaners, and we'll probably all breathe a sigh of relief when those jobs are finally automated, but will we stop wanting to make art or music when AI can make it "just as good"? Even in more practical fields than art or music, it's more likely that we'll see AI-human collaborations rather than AI doing everything by itself without human input. What is human is constantly in flux anyway. Even if we consider just technology's effect on the human, there are humans now with all sorts of prosthesis that make them "more than human" in some ways, and this trend itself will likely continue, with things like more and more advanced neural interfaces, memory upgrades, sensory enhancement, etc. The future of humanity will likely be as some sort of cyborg.
    - IntrepidWorm 5y ago
      I think back to Calhoun's rat utopia experiments and wonder if humans could ever be happy in a post-work society. https://www.smithsonianmag.com/smart-news/how-mouse-utopias-1960s-led-grim-predictions-humans-180954423/ https://www.smithsonianmag.com/smart-news/how-mouse-utopias-... While the predictions made by the study regarding human behavior have since been called into question (and because a human version of these experiments would be horribly unethical), it does bear to think about what the world would look like if all work was performed by automated means free of human interaction. On one hand, there are many jobs that would be seldom missed by most people. On the other, people can find fullfillment in just about anything.