Install
Please confirm you are human
This browser or connection looks automated. Press and continuously hold the control for 3 seconds to enable Google-hosted web results and, when separately allowed, AI-assisted answers.
A successful check enables 100 search requests. Interactive access does not authorize scraping, systematic collection, or reuse of search output.
News
🧠 I Trained a Massive Word2Vec Model on 13 Billion Russian Fiction Words — Here’s What Happened
1+ week, 7+ hour ago (258+ words) TL;DR: I built a lemma‑based Word2Vec model (CBOW, 300d) on a huge corpus of Russian fiction (13B tokens → 7.3B after cleaning). You can load it with Gensim and explore semantic neighborhoods of words like слово, язык, речь. The model captures literary semantics…...
Fine-Tuned Lie Detectors Failed to Generalize
3+ week, 2+ day ago (1577+ words) Research done as part of MATS and the Anthropic Fellowship. Misalignment is hardest to correct when it's concealed: a model that pursues the wrong goals or holds false beliefs in the open gives operators something to act on. This is…...
Our Documentation Was Lying. The Model Believed It.
3+ week, 5+ day ago (1544+ words) There is a variable called slp — sea-level pressure. Our API documentation lists it as available. Our error messages list it among the valid options. Our fine-tuned model, asked about air pressure over Belgrade, will happily write you fifteen lines of…...
AI has the potential for harm just as it has the potential for good, says Richard Clarke
1+ mon, 1+ day ago (35+ words) CNBC AI has the potential for harm just as it has the potential for good, says Richard Clarke...
The most dangerous question in the AI age - AOL
1+ mon, 4+ day ago (1640+ words) The labor debate is a decoy. The real battle is who controls attention, criteria, and the right to say no. The question “Where do we put the humans?” isn’t pragmatic. It’s the survival-world confessing its anthropology. There is a sentence…...
My Fall-Detection Model Scored 94%, and It Was Lying to Me
1+ mon, 1+ week ago (1025+ words) How a single evaluation choice inflated my results by 25 points, and what rebuilding honestly taught me about ML systems people might depend on My fall-detection model scored 94.3% accuracy. I went back over the terminal output and the confusion matrix more…...
The Attack That Waits Three Weeks Before Doing Anything
1+ mon, 1+ week ago (217+ words) An injected instruction sat in a test memory store for nine sessions on average before anyone — human or model — noticed it. Independent …...
Existential Risk from AI: An Exposition for Mathematicians — LessWrong
1+ mon, 1+ week ago (9+ words) Discuss Existential Risk from AI: An Exposition for Mathematicians...
The Human Side of AI — Part 1
1+ mon, 1+ week ago (37+ words) AI Safety Has Been Asking the Wrong Question We’ve spent years trying to make AI safer. We haven’t spent nearly enough time asking whether AI is changing the …...
Mathematical Movement of the Year: The MIMC 2026 Olympiad Opened Powerfully in Ulaanbaatar » News of Kyrgyzstan, Bishkek and Osh — latest events today
1+ mon, 2+ week ago (35+ words) Mathematical Movement of the Year: The MIMC 2026 Olympiad Opened Powerfully in Ulaanbaatar open.kg - Traditional vibe: powerful folk songs and colorful national dances. - Steppe sound: the live, hypnotic sound of the legendary morin khuur....