Though chatbots can now seemingly pass the Turing Test - with certain provisos - you might be happy to hear that Generative AI on its own can't beat us at chess. Or rather, it can beat me, but it wouldn't consisitently win against an experienced amateur player, and it probably would lose - at least most of the time - to a pro. This is because the Large Language Model (LLM) that makes it work means it only has access to textual records of games already played and notated in standard Chess notation, which looks something like this: Nf3 Nc6. As I understand it, it has thousands (or millions) of notated games that have appeared in books and newspaper chess columns and so on, and it sifts them looking for moves that were successful when other moves preceding them looked similar or identical to the moves a human player is making in a game with it! (I can't even imagine how much energy this must eat up.)
In October, 2025, I asked GPT a question similar to the one I'd asked it back in 2022 about whether individual knowledge is still valuable. It answered at greater length in more nuanced ways in 2025. I'm not sure if this is because the model has improved or because I've spent time with it lately and it's gotten to "understand" what sort of an answer I might like and can handle. Here's what it "thinks" about Individual Knowledge in 2025:
Short answer: yes — individual knowledge still matters, though the why is shifting.
1. Cognitive scaffolding
What you hold in your “meat mind” isn’t just storage; it’s structure.
Personal knowledge forms the scaffolding that lets you interpret, question, and reframe whatever comes from outside sources (like me). If you outsource too much of that to external systems, you lose the internal architecture that makes thought coherent — you can retrieve information, but not understand it.
2. Embodied and emotional anchoring
Knowledge in a human isn’t purely abstract — it’s bound to sensory experience, emotional tone, and context.
That embodied quality makes memory sticky and meaningful. A fact tied to lived experience becomes intuition; a fact recalled from a database stays inert. So internal knowledge supports judgment and empathy, not just recall.
3. Epistemic independence
When individuals stop cultivating internal knowledge, collective understanding becomes fragile.
If everyone relies on the same external systems (say, AI retrieval or shared platforms), the range of independent interpretation collapses. Human knowledge diversity — the capacity for unexpected insight or resistance — depends on individually maintained mental ecosystems.
4. That said…
You don’t need to memorize everything. What matters is keeping an active, organized cognitive core: enough internal knowledge to reason, to test claims, to form taste. The rest can indeed be distributed across notes, devices, or AIs — as long as you remain the integrator, not the echo.
So yes: the value of personal knowledge isn’t in hoarding information, but in maintaining a living cognitive style.
Would you like me to sketch how that idea might play out differently for a student versus a teacher?
I didn't ask it. You always can.