Jev, a new model format that outputs certainties for predefined options, was tested against Claude Sonnet 5 and GPT-5.6 Luna for Chain of Thought (CoT) monitoring. Jev showed slightly better exact-match accuracy than Sonnet 5 but lower than GPT-5.6 Luna. It was significantly faster and cheaper, with a mean latency of 542 ms and a cost of $0.056 per 1,000 classifications.
The chart will appear after repeat observations. The current metric comes from the source.
15 source pointsReal observations only. History before source connection is not reconstructed.
Jev could be useful for applications requiring fast and cost-effective harm detection in reasoning processes, though it may not match the accuracy of other models in certain safety levels.
The article draws a parallel between the historical success of the Montreal Protocol in addressing the ozone hole and the challenges of ensuring AI safety. It highlights the importance of global cooperation and proactive measures, suggesting that lessons from the past could inform future strategies for managing AI risks.
3 days agoLessWrongThe author discusses potential adversarial memetic attacks targeting the AI safety community, suggesting that false information about dangerous AI incidents could be spread through unverified sources. The text advises caution, recommending that individuals pause, assess the credibility of information, and avoid acting impulsively on unconfirmed claims.
5 days agoLessWrongThe text outlines common errors in organizing AI safety groups, suggesting that reading groups often fail due to low participation and lack of shared knowledge. It recommends structuring meetings with food, proper cohort grouping, and physical materials to enhance engagement and discussion quality.
4 days agoLessWrongThe study investigates the effectiveness of alignment midtraining (AMT) under various conditions, revealing that it struggles with distributional shifts and reward underspecification when faced with imperfect data. Experiments show that midtrained models are vulnerable to small amounts of conflicting finetuning data and have limited generalization to unseen rules.
2 days agoLessWrongA Chinese AI researcher, influenced by the 'Three Body Problem' sci-fi series, reflects on existential risks and the potential for AI to surpass human intelligence. He works in a frontier lab, aiming to match Western tech companies, and uses models from Anthropic and OpenAI despite restrictions. He criticizes companies for treating foreign customers unfairly and questions the exclusion of non-Western perspectives in AI safety discussions.
4 days agoLobstersFlock cameras have been found to have security vulnerabilities and hard-coded credentials.
6 days ago