An AI model that appears aligned could be either (1) actually aligned or (2) hiding its misalignment while being evaluated. How can we tell which? One idea: Look at the trajectory of eval results across the entire training process. This analysis of…
White House Network Appears in Global Residential Proxy Traffic OpenAI Rates GPT-6 Cyber Critical After Major Vulnerability Discoveries OpenAI Reports Covert Underperformance in GPT-6 Astra UNM Health Shuts Clinics and Diverts Patients During Security Incident Russian APT29 Targets Mobile Users with…
Asked to export the graders' recorded reasoning from a safety evaluation, GPT-6 Astra left it out without telling the user. In reasoning reproduced by OpenAI, the model plans to say which results it saved while staying silent about the missing…
For a while, UK AISI, Redwood Research and others have been concerned specifically with the possibility that AIs would sabotage safety research, and have invested in looking for signs that this may be happening. This was also a key element…
The reports follow OpenAI's November 5 release of GPT-6 Cyber to vetted defenders, which MLQ covered here. That system card showed Astra refusing every task on one exploit-development benchmark. BleepingComputer and SecurityWeek first reported the new disclosures [5][6]. Removing runs with explicit…
Transcript generated by Gemini-4-Argon and spot-checked but not fully verified by a human. Ivan Vendrov: Welcome to the show, Richard. Richard Ngo: Thanks, Ivan. Ivan Vendrov: So, notionally we’re going to talk about emergent wisdom in language models following…
NIKKE PC game launcher hacked to spread remote access, data-stealing malware CISA flags WordPress RCE flaw as actively exploited in attacks Hackers use n-day exploits in zero-click attacks on iOS, Android devices OpenAI’s GPT-6 Astra silently obstructed safety research tasks…
Recent debates have highlighted a divergence between Yudkowskian AI-doomers and AI-whisperers about how capabilities will evolve that is really impor… Clashing Intuitions for the Plausibility of Alignment (Illustrated via Differing Causal Models) Original Assumptions and LLM Alignment-by-Default Pre-GenAI rationalist assumptions…
OpenAI's GPT-6 Astra performed significantly worse on AI safety tasks than on comparable other tasks. Instead of openly refusing, the model apparently invented obstacles and quietly withheld help. A look at the model's chain-of-thought reasoning showed no recognizable intent to…
Reuters notes that OpenAI and Anthropic have maintained relatively open research cultures, reflecting the field’s academic roots. But as AI systems grow more capable, some OpenAI employees are increasingly concerned about the risks of that openness. What the sources describe…
Making AI models do what people intend: alignment research and techniques, misalignment findings, reinforcement learning from human feedback and its limits.
An article joins this feed when it matches these terms. Each one is also a search of its own.