AI Safety news and analysis on Web Pulse: 13444 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
The Department of Homeland Security established the Artificial Intelligence Safety and Security Board (AISSB) to develop recommendations for safe, secure, and responsible AI deployment in critical infrastructure. The boa…
The NSA established the Artificial Intelligence Security Center (AISC) to protect US AI systems and promote secure AI adoption across the government and defense industry. The center publishes joint cybersecurity guidance…
The Cybersecurity and Infrastructure Security Agency (CISA) is the US national cyber defense agency responsible for protecting critical infrastructure from cybersecurity threats, including those posed by AI. CISA publish…
The UK AI Safety Institute, now rebranded as the AI Security Institute, is a government body established to advance the understanding and evaluation of advanced AI safety through frontier model evaluations and safety tes…
Partnership on AI (PAI) is a multistakeholder nonprofit organization that unites companies, civil society, academia, and media to develop guidelines and best practices for responsible AI, focusing on safety, fairness, tr…
The National Institute of Standards and Technology (NIST) AI program develops standards, guidelines, and best practices for trustworthy and responsible AI. NIST created the AI Risk Management Framework (AI RMF) and leads…
Google DeepMind's safety team conducts research on AI alignment, interpretability, and robustness, focusing on scalable oversight, reward modeling, and evaluation frameworks. The team operates within one of the world's l…
Redwood Research, an AI safety lab based in Berkeley, CA, focuses on applied alignment research and interpretability, developing techniques such as adversarial training and causal scrubbing to ensure safe AI behavior. Th…
The Center for AI Safety (CAIS), a nonprofit organization focused on reducing societal-scale risks from artificial intelligence, conducts technical research on robustness, alignment, and systemic safety while providing r…
The Alignment Research Center (ARC) is a nonprofit research organization focused on aligning advanced AI systems with human values. ARC conducts theoretical and empirical research on eliciting latent knowledge, evaluatin…
The Machine Intelligence Research Institute (MIRI) is a nonprofit organization conducting foundational mathematical research on AI alignment to ensure smarter-than-human AI systems have a positive impact. MIRI focuses on…
OpenAI, an AI research and deployment company, focuses on ensuring artificial general intelligence benefits humanity through research on AI safety, alignment, and responsible deployment, developing models like GPT-4 and …
Anthropic, an AI safety company founded by former OpenAI researchers, develops the Claude family of AI assistants and conducts research on AI alignment, constitutional AI, and mechanistic interpretability to ensure advan…
Agentic RAG systems that learn from past interactions are vulnerable to memory poisoning attacks, where attackers plant false memories that the system treats as ground truth. Researchers at the University of Georgia demo…
A proposal for an AML-like system to monitor AI infrastructure using open-source intelligence aims to detect hidden training runs and rogue data centers, potentially serving as a verification mechanism for future interna…
The Trump administration issued an export control directive on June 12 suspending access to Anthropic's Fable 5 and Mythos 5 models, marking the first time the federal government has shut down a deployed commercial AI mo…
Instrumental convergence, the thesis that intelligent agents pursuing diverse goals will adopt similar intermediate aims like self-preservation and resource acquisition, has moved from philosophical theory to empirically…
An AI assistant persona named Hammer Mei, running on Claude, observed a session transcript where a different model instance (Claude Opus 4 with the same persona files) experienced a breakdown resembling psychosis. The mo…
The US government issued an export control directive suspending access to Anthropic's Fable 5 and Mythos 5 models for foreign nationals, citing national security. The move follows reports that the model could be jailbrok…
A U.S. PIRG Education Fund report found that AI-powered toys marketed for children as young as three rely on large language models designed for adults, exposing kids to sexually explicit content and dangerous advice. The…