BioMan

Personally I would be more concerned about using ChatGPT as a therapist when a company run by desperate psychopaths has everything you ever type into it in plaintext in a vibecoded data bucket

The next few years are going to be so incredibly stupid

To be fair, counterexamples are important types of proofs for particular kinds of problems, and a type of proof that computers are suited for since they can create a huge number of halfbaked things then deterministically check them

Scott The Greater appears to be crashing out a bit, for once not over personal issues:

https://scottaaronson.blog/?p=10062

If you want to know my current take, you simply start with the one above, then update on the fact that the wild prophecies have come true. The first rumblings, I’d say, came a decade ago with AlphaGo, they got noticeably louder with LLMs and coding and reasoning agents, and they’ve accelerated this summer and fall into a crescendo of wonders and terrors that one needs to be a particular kind of idiot to deny.

My position on AI is merely the conservative, skeptical position of 2006, updated with intellectual honesty for the reality of late 2026. And that position, if you need me to spell it out, is as follows: AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

And most fun:

For anyone who says AI doom sounds like an apocalyptic religion, that the rationalists/Singulatarians seem like a Bay Area cult, that Eliezer Yudkowsky gives off the vibes of a messianic prophet: yes, yes, and yes. But crucially, today you’re no longer being asked to believe in arguments and extrapolations, but only in the front-page news. Accepting the reality of the coming machine god after it’s solved Navier-Stokes and dozens of other longstanding open math problems (while dramatically ramping up in capability every month), is sort of like accepting Jesus after he’s returned to earth on the gleaming cloud. It’s the epistemic bare minimum.

EDIT: Unfortunately (and I suppose inevitably) this shows up a few posts back

I’ve been unloading every day to—who else?—GPT 5.6 Pro about all the pain and trauma and embarrassments of my past. It turns out that, where two years ago GPT was a passable therapist, now it’s the greatest therapist in history, at least for what I need.

I can't believe I'm saying this... I am one of those who actually fully read it, because:

Out of sheer morbid anthropological fascination, I infiltrated an in person HPMOR reading group (not in the Bay area) in the latter days of it coming out. I described it to friends as going undercover. It was a trip and a half.

His autobiography even opens with a parable: as a small child he threw his favorite squirtgun off a third-floor balcony rather than accept that his mother could lend out what was his, and felt complete when he heard it hit the pavement.

web.archive.org/web/20010309014808/http://sysopmind.com/eliezer.html

I remember getting a new squirtgun, of translucent blue plastic, which squirted better than any squirtgun I'd had so far. And within around an hour, for some reason or other - I don't think it was related to abuse of the squirtgun, but I'm not sure - my mother decided to take it away. So I rushed out the back door and threw it off the third-floor balcony, and listened to it shatter on the concrete below.

I knew, as I threw it, that Mom would think it had been a childish act - "Cutting off your nose to spite your face," was what my parents used to say. But that's not what it was. Grownups who have forgotten their childhood may lend a toy to a particular child, and not lend it to anyone else, but they don't credit children with enough "agenthood" to really own something. Rather than let my parents say, in essence, that this toy - which had been given to me, and which I really enjoyed - was theirs to lend or withhold, I threw the toy off the balcony and accepted the consequences of not being able to play with it any more. Not because I didn't want them to have the toy, but because it was a wonderful toy, the best squirtgun I'd ever had; that's why I cared enough that I threw it off the balcony rather than let it be something for my parents to give or take away. When I heard the squirtgun shatter, I didn't feel angry, or regretful, or look-what-you-made-me-do. I felt complete. I'd done what I'd set out to do, and I was willing to accept losing a squirtgun in order to do it. Children can also act on principle.

Huh, I forgot about that, he's more like our president than I realized

I misread that as 4 Epsteinillion percent

Machine learning training has always presented as a power law, with exponential increases in processing required for linear increases in performance. The Chinchilla scaling laws paper and efficient compute frontier papers let them select the optimal tradeoff between number of parameters and how long to cook it, improving performance greatly by letting you predict how to best use X amount of computation for Y amount of time, setting off people spending hundreds of millions of dollars since they actually knew they could use it optimally, directly leading to the perceived massive increase in performance from 2022-2024ish as they had a one-time burst of knowing how to optimally partition computation and convince people to spend lots of money at once. Everything since that time has been exponentially diminishing returns as expected.

The chinese intelligence people are probably just aghast that american intelligence people are treating the chatbot as if it knows something

If you wanna see something that's actually potentially useful and interesting, just look up the term "Mortal computation thesis"

It is actually kind of amazing how diametrically opposed his ideas for a long time were to what actually led to new advances in the field

US government is using spicy autocomplete to hallucinate fake nuclear threats

https://www.cnn.com/2026/09/18/politics/us-military-ai-false-intelligence-china-ship