Join free with your email
Free. No spam, ever — we'll never share your email address and you can opt out at any time. Já tem conta? Entre
Tracked through public AI activity and peer connections inside the directory.
SpaceX-owned xAI, the developer of Grok, sued Minnesota AG to challenge a law coming into effect in the state that will ban "nudify" apps, use of these AI image generators to make non-consenting intimate images, aka AI deepfake porn of real people including minors. www.cnbc.co…
this is one of the most insane fucking things i have ever read www.amnesty.org/en/documents...
i think pre-planning rhymes is the best example of this behavior, where the model preactivates features associated with words which complete the rhyme and "stores" them as annotations on the newline token. transformer-circuits.pub/2025/attribu...
anyway, if you're interested in Hutterian models of AI, here's Hutter (2015) and his paper about why the MIRI model of the Bad Robot doesn't work arxiv.org/abs/1510.04931
Philosopher: "In my book I invented Color Scientist Mary as a thought experiment." Tech Company: "At long last, we have created Color Scientist Mary from classic philosophical paper What Mary Didn't Know (1986)" arxiv.org/abs/2109.06129
i think there's an actual question embedded in the "if Claude were representing Genghis Khan and Julius Caesar, would we say that it was really the thing itself?" hypothetical, and it is "should an actor playing Brutus be permitted to stab Caesar to death on stage."
i don't know what you mean by "top layer," but the answer is "this is not just possible in principle, but there are preliminary results on precisely this question." (i do not think that small-model results necessarily generalize to 'this is how every model does it.') arxiv.org…
galaxy brain: a LLM is ffmpeg arxiv.org/abs/2306.04050
www.wired.com/story/ai-pr-...
siliconvalleynotes.substack.com/p/agi-cnc-an...
Apparently in the next version of Fable, Anthropic changed their tokenizer based on real world usage and made “Let me check rather than answer from memory” into a single token, reportedly saving 43% of all token generation very interesting tinyurl.com/58zf9d3c
the entire research group corresponding to my operations group was redeployed to support a project which is a priori impossible and which had no real mathematical affinity to their research area techcrunch.com/2026/06/24/a...
i think the closest analogy to 'what is going on at AI companies' is 'what was going on at JPL in the 1950s,' where communists, Satanists, swingers and adherents of a bizarre Russian Orthodox-adjacent techno-optimist microsect ended up doing a tremendous amount of important rocket science
it's probably better to trust the experts on the topics of their expertise, except people who do machine learning, who have all gone insane
using LLMs is depriving me of the basic joys of human existence,,,, such as writing SQL queries by hand
the problems with "breaking the AI safety cartel" is that virtually everyone who wants to do that was saying that AI didn't work six months ago, and that for all their bad policy ideas most of the world's technical knowledge about reinforcement learning is at least in its membership's social circle
papers: so the fundamental unit of computation in a neural network is the 'layer' me: oh great. could i get the definition of 'layer' papers: oh heavens no
i very much do not like the current idea that it is ridiculous to ask questions about AI which were perfectly reasonable to ask in 2019
the reason that so many people in the field are attracted to Rationalism is that they were thinking about AI before it was on the horizon. a lot of people got into the field because they were excited about this.
instead of AI we should have invented books with wider margins so that our marvelous proofs could fit there
altman's "we have to accept some bad things will happen" stuff is what he tells the SVPs about why OpenAI's security posture is essentially "let the models do whatever they want."
oh huh i can ... hm. i don't know if the entire ya-Sattra thing will work from a ChatGPT project, but it turns out i can share it out? does anyone want this?
Center = rev. howard arson. Left = members they follow (green edges). Right = members who follow them (blue edges). Top = mutual follows (orange edges, slightly larger). Drag any node to reposition; click to open that profile.
Add the Who’s Who of AI badge to your site or bio. It links back to this profile.
Markdown: [](https://aiweekly.co/whos-who/person/theophite-bsky-social)