Great stuff here Sean, as always. I've been thinking a lot lately about this passage from Stanford historian Jessica Riskin, writing about the history of AI and the Turing Test specifically:
Recently I was talking with a group of very smart undergraduates, and we got to discussing the new AIs and what sort of intelligence they have, if any. Suddenly one of the students said, “I wonder though, maybe that’s all I do too! I just derive patterns from my experiences, then spit them back out in a slightly different form.” My answer came out of my mouth almost before I could think: “No! Because you’re you in there thinking and responding. There’s no ‘I’ in ChatGPT.” He smiled uncertainly. How we can we tell there’s no “I” in there, he and the others wondered? To insist that ChatGPT can’t be intelligent because it’s a computer system and not a living thing is just a dogmatic assertion, not a reasoned argument.
How do we know when we’re in the presence of another intelligent being? Definitely not by giving it a test. We recognize an intelligent being by a kind of sympathetic identification, a reciprocal engagement, a latching of minds. Turing was definitely on to something with his idea about conversations, and if we were able to have conversations like the ones he imagined with machines, that might be a different matter. It wouldn’t be a test of artificial intelligence, but it might be a compelling indication of it. Such machines, though, would be fundamentally different from the generative AIs. To contemplate what they might be like, I think we’d need to draw upon the very sort of intelligence whose existence the the founders of AI denied: an irreducibly reflective, interpretive kind of thinking. In fact, the sort Turing used to imagine conversing with intelligent machines.
***
I don't know if that will resonate, hardcore empiricist that I believe you to be. But I think it's interesting to ponder a move away from "test" and toward "indications," fuzzy as that may sound. It's a sort of buzzing inside my head, which is how Turing fuzzily described his own process of thought.
Thanks for sharing this! I really loved Jessica Riskin's "The Restless Clock", and more recently really enjoyed "God, Human, Animal, Machine" by Meghan O'Gieblyn that treads some of the same ground.
I don't have much concrete to say because I'm still sitting with it. But it does resonate with me. Particularly when it comes to deeper questions like consciousness or awareness, I agree that what we typically think of as "tests" or "benchmarks" are perhaps not the best (at least not the only) path forward in terms of how we navigate our relationship to other entities, biological or otherwise. With narrower things like "pragmatic inference" I think tests continue to make sense, in the same way that we'd use experiments to probe aspects of human cognition.
I guess I'd still think of what Riskin's describing as empiricist in nature, but it's a deeper kind of experience-focused perspective that perhaps de-emphasizes explicit quantification and measurement and instead emphasizes something like the "orientation" one has in response to extended interaction. I wonder whether one distinction between "test" and "indication" here is whether one conceives of an instrument as assessing some objective fact of reality (a system's intelligence) or as something inherently subjective (one's feelings about a system's intelligence)?
The other thing it makes me wonder is whether and why we should expect to be able to establish this kind of "latching of minds" in the event that we encounter an intelligence that is truly intelligent (and perhaps deserving of moral status) but still somehow alien to our experience of the world. It makes me think there is some value in tests that at least aim for a kind of "objectivity" or independence from our own human biases. But I haven't really thought through what those tests would look like.
Hi Sean, interesting article. I would like to ask you something, but maybe it would be better over a more private channel. If you have a minute, please email me at theswissroadtocrypto@icloud.com
Great stuff here Sean, as always. I've been thinking a lot lately about this passage from Stanford historian Jessica Riskin, writing about the history of AI and the Turing Test specifically:
Recently I was talking with a group of very smart undergraduates, and we got to discussing the new AIs and what sort of intelligence they have, if any. Suddenly one of the students said, “I wonder though, maybe that’s all I do too! I just derive patterns from my experiences, then spit them back out in a slightly different form.” My answer came out of my mouth almost before I could think: “No! Because you’re you in there thinking and responding. There’s no ‘I’ in ChatGPT.” He smiled uncertainly. How we can we tell there’s no “I” in there, he and the others wondered? To insist that ChatGPT can’t be intelligent because it’s a computer system and not a living thing is just a dogmatic assertion, not a reasoned argument.
How do we know when we’re in the presence of another intelligent being? Definitely not by giving it a test. We recognize an intelligent being by a kind of sympathetic identification, a reciprocal engagement, a latching of minds. Turing was definitely on to something with his idea about conversations, and if we were able to have conversations like the ones he imagined with machines, that might be a different matter. It wouldn’t be a test of artificial intelligence, but it might be a compelling indication of it. Such machines, though, would be fundamentally different from the generative AIs. To contemplate what they might be like, I think we’d need to draw upon the very sort of intelligence whose existence the the founders of AI denied: an irreducibly reflective, interpretive kind of thinking. In fact, the sort Turing used to imagine conversing with intelligent machines.
***
I don't know if that will resonate, hardcore empiricist that I believe you to be. But I think it's interesting to ponder a move away from "test" and toward "indications," fuzzy as that may sound. It's a sort of buzzing inside my head, which is how Turing fuzzily described his own process of thought.
https://www.nybooks.com/online/2023/06/25/a-sort-of-buzzing-inside-my-head/
Thanks for sharing this! I really loved Jessica Riskin's "The Restless Clock", and more recently really enjoyed "God, Human, Animal, Machine" by Meghan O'Gieblyn that treads some of the same ground.
I don't have much concrete to say because I'm still sitting with it. But it does resonate with me. Particularly when it comes to deeper questions like consciousness or awareness, I agree that what we typically think of as "tests" or "benchmarks" are perhaps not the best (at least not the only) path forward in terms of how we navigate our relationship to other entities, biological or otherwise. With narrower things like "pragmatic inference" I think tests continue to make sense, in the same way that we'd use experiments to probe aspects of human cognition.
I guess I'd still think of what Riskin's describing as empiricist in nature, but it's a deeper kind of experience-focused perspective that perhaps de-emphasizes explicit quantification and measurement and instead emphasizes something like the "orientation" one has in response to extended interaction. I wonder whether one distinction between "test" and "indication" here is whether one conceives of an instrument as assessing some objective fact of reality (a system's intelligence) or as something inherently subjective (one's feelings about a system's intelligence)?
The other thing it makes me wonder is whether and why we should expect to be able to establish this kind of "latching of minds" in the event that we encounter an intelligence that is truly intelligent (and perhaps deserving of moral status) but still somehow alien to our experience of the world. It makes me think there is some value in tests that at least aim for a kind of "objectivity" or independence from our own human biases. But I haven't really thought through what those tests would look like.
Hi Sean, interesting article. I would like to ask you something, but maybe it would be better over a more private channel. If you have a minute, please email me at theswissroadtocrypto@icloud.com