Philosophy of Mind Cannot Explain Artificial Intelligence
Benjamin Wallace’s New York Times article, “The Revenge of the Philosophy Majors,” demonstrates that the discipline of philosophy of mind has become indispensable to artificial intelligence (AI), but it also betrays how the Western understanding of consciousness is like a fish that cannot see the water in which it swims. The article profiles philosophers now working at Anthropic, DeepMind and affiliated nonprofits, presenting their rise as evidence that the technology industry has begun to take consciousness, ethics and the nature of mind more seriously. But that conclusion is too quick. What the article unintentionally exposes is a deeper misguidedness. It tries to demonstrate AI’s "consciousness" by asking Claude to debate who is the best Beatle when a better question would be asking, "How is the universe operating so that Paul McCartney woke up one day with the entire melody of 'Yesterday' floating in his head?"
Amanda Askell is perhaps the clearest avatar of this moment. Her work at Anthropic, especially around Claude’s "character" and constitutional alignment, is plainly serious and influential. Yet her prominence also highlights the limitations of the current framework. If the project is genuinely to understand consciousness, moral life and the structure of human thought, then Western philosophy alone is an inadequate guide. The problem is that the discourse has been too narrowly organized around a set of assumptions that pass themselves off as neutral when they are anything but.
Philosophy of Mind
Like Amanda, I studied philosophy of mind in graduate school, but it was only years later in another graduate program studying Buddhism with Alan Wallace and then through years as a clinical psychotherapist treating patients that I came to believe that Western philosophy may be asking the wrong questions. Philosophy of mind tends to ask what consciousness "is" in abstraction, as though one might define it with the sort of precision one brings to problems of logic. But the consciousness that created Mahler's 2nd symphony and Rothko's totems is not awaiting a scientific formulation or calculation. It is a lived field of experience: embodied, affective, unstable, historically mediated, resistant to the very language used to describe it and… profoundly and irreducibly irrational and non-linear. Aside from supposed AI "hallucinations," large language models (LLMs) do not possess these all-too-human talents.
Please select this link to read the complete article from Psychology Today.