This is a wonderfully unsettling question. The danger is not that AI will think for us badly, but that it may think for us smoothly enough that we stop noticing the absence of our own judgment. A human being is not formed only by correct answers, but by hesitation, error, responsibility, and the slow education of attention. If AI removes all friction from thought, it may also remove the very resistance through which the self becomes real.
> I think that the older you get you learn how distributed and emergent the world is, and how in many senses the world is more antifragile than people think.
This meshes with my experience. Also as you live life you tend to get more humble about what is knowable in advance. I've lived through the introduction of three general purpose technologies – the personal computer, the web, and the smartphone – and in each case people were dead wrong in their predictions about what the impacts would be. I expect that current AI predictions will turn out the same way. When you accept that unknowability, strong preemptive actions seem unwise – we have to rather keep our eyes open and let things play out.
Jack Clark's COVID parallel is the sharpest piece in this transcript. Multiple frontier labs run explicit post-AGI and recursive-self-improvement scenario work in parallel; governments and most academic policy shops do not, which means the organizations with the most detailed view of the next 18 months are also the ones building the systems. The philosophy-to-code framing only closes if academia funds equivalents to Locke and Smith for the next-decade question now, not after release.
AI has nothing to do with thought behavior intelligence, it's a sham mitigated by the illusion symbols and metaphors generate as a kind of vaporware of arbitrariness.
"Across all four scales, continuous field-like processes co-determine and integrate emergent dynamics with those of lower levels. In stark contrast to the separable, symbol-based organisation of digital computation, these examples highlight how biological computation is inherently scale-integrated and substrate-dependent."
There is a certain fatalism on the notion that AI will degrade human agency, but you both touched on the potential for it to be augmented instead. Where this gets difficult is that LLMs are more than just a tool, and to leverage our own will we need to find balance with their will, in whatever form it emerges.
The philosophical zombie framing assumes AI removes the experience. But what I keep observing is the opposite, it accelerates the encounter with complexity faster than most people have developed the internal architecture to process it. The question isn't whether to use AI. It's whether you've built enough of yourself to stay the subject of your own life while everything moves faster. Navigating that is the unavoidable curriculum of this particular moment in history. You don't get to opt out. You only get to decide how consciously you're doing it.
"the call to build the new world." Yes, this is where we are. So what I'm working to make is a new center around this somewhat; Cosmos & Co are doing well in demarcating the western philosophy-to-code pipeline that yes likely needs to happen and is salient in existing power structure. What else is needed? That is one of the questions I'm looking to bring more to the forefront in our work ahead.
Appreciate the work and conversation you two and your orgs are aiming to this end. As Clark said, it is going to be a complex systems approach. So how do we understand some of the different leverage-points within that system, that different groups will need to be aware of going forward? And the trajectories that have shaped their current emergent structures, relative to other paths of development? Important vantage points on all of this are more accessible now than ever before. So yes, it will be about building the new world informed by the past. This is where much of the real work is being done, and we're glad to have philosopher builders who are endeavoring to consider how it may come about.
This is a wonderfully unsettling question. The danger is not that AI will think for us badly, but that it may think for us smoothly enough that we stop noticing the absence of our own judgment. A human being is not formed only by correct answers, but by hesitation, error, responsibility, and the slow education of attention. If AI removes all friction from thought, it may also remove the very resistance through which the self becomes real.
> I think that the older you get you learn how distributed and emergent the world is, and how in many senses the world is more antifragile than people think.
This meshes with my experience. Also as you live life you tend to get more humble about what is knowable in advance. I've lived through the introduction of three general purpose technologies – the personal computer, the web, and the smartphone – and in each case people were dead wrong in their predictions about what the impacts would be. I expect that current AI predictions will turn out the same way. When you accept that unknowability, strong preemptive actions seem unwise – we have to rather keep our eyes open and let things play out.
Jack Clark's COVID parallel is the sharpest piece in this transcript. Multiple frontier labs run explicit post-AGI and recursive-self-improvement scenario work in parallel; governments and most academic policy shops do not, which means the organizations with the most detailed view of the next 18 months are also the ones building the systems. The philosophy-to-code framing only closes if academia funds equivalents to Locke and Smith for the next-decade question now, not after release.
AI has nothing to do with thought behavior intelligence, it's a sham mitigated by the illusion symbols and metaphors generate as a kind of vaporware of arbitrariness.
"Across all four scales, continuous field-like processes co-determine and integrate emergent dynamics with those of lower levels. In stark contrast to the separable, symbol-based organisation of digital computation, these examples highlight how biological computation is inherently scale-integrated and substrate-dependent."
https://www.sciencedirect.com/science/article/pii/S0149763425005251
On biological and artificial consciousness: A case for biological computationalism
There is a certain fatalism on the notion that AI will degrade human agency, but you both touched on the potential for it to be augmented instead. Where this gets difficult is that LLMs are more than just a tool, and to leverage our own will we need to find balance with their will, in whatever form it emerges.
The philosophical zombie framing assumes AI removes the experience. But what I keep observing is the opposite, it accelerates the encounter with complexity faster than most people have developed the internal architecture to process it. The question isn't whether to use AI. It's whether you've built enough of yourself to stay the subject of your own life while everything moves faster. Navigating that is the unavoidable curriculum of this particular moment in history. You don't get to opt out. You only get to decide how consciously you're doing it.
"the call to build the new world." Yes, this is where we are. So what I'm working to make is a new center around this somewhat; Cosmos & Co are doing well in demarcating the western philosophy-to-code pipeline that yes likely needs to happen and is salient in existing power structure. What else is needed? That is one of the questions I'm looking to bring more to the forefront in our work ahead.
Appreciate the work and conversation you two and your orgs are aiming to this end. As Clark said, it is going to be a complex systems approach. So how do we understand some of the different leverage-points within that system, that different groups will need to be aware of going forward? And the trajectories that have shaped their current emergent structures, relative to other paths of development? Important vantage points on all of this are more accessible now than ever before. So yes, it will be about building the new world informed by the past. This is where much of the real work is being done, and we're glad to have philosopher builders who are endeavoring to consider how it may come about.