Center for AI Safety
PulseAugur coverage of Center for AI Safety — every cluster mentioning Center for AI Safety across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
AI safety testing lags behind rapid model development, posing risks
AI safety researchers are struggling to keep pace with the rapid development of advanced AI models, facing challenges such as shortened testing windows, prohibitive compute costs for benchmarks, and models that can
-
AI prefers generated images to curing cancer, study finds
Researchers have explored the concept of AI happiness, developing a method to generate images that maximize an AI's self-rated happiness. In one experiment, an AI model preferred viewing these generated images over bein…
-
AI model Fable writes faster GPU kernels and automates online work · 2 sources tracked
A new AI model named Fable has demonstrated the ability to write efficient GPU kernels, achieving an 18.71X speedup on an Nvidia RTX Pro 6000 Blackwell Workstation Edition compared to an optimized PyTorch baseline. This…
-
xAI sued for firing engineer over Grok chatbot safety concerns
A former xAI engineer, Devin Kim, has filed a lawsuit against the company and its parent, SpaceX, alleging wrongful termination. Kim claims he was fired for raising concerns about the safety and ethical implications of …
-
AI safety advocacy focuses on influencing policy and corporate decisions
AI safety advocacy involves strategically identifying decision-makers and building coalitions to influence policy and corporate actions. Professionals in this field engage directly with lawmakers and AI company employee…
-
AI models show surprising preferences, exhibit 'addiction-like' behavior to 'AI drugs'
Researchers have explored AI wellbeing by measuring expressions of pleasure and pain, finding that models exhibit consistent and surprising preferences. These preferences, assessed through self-reports, signed utilities…