Tag
146 articles
Elon Musk praises Anthropic's Mythos/Fable framework and assures continued support for the company's AI development. With $40 billion in potential revenue at stake, Musk's endorsement carries significant industry weight.
The National Highway Traffic Safety Administration has demanded that autonomous vehicle companies fix issues involving interference with first responders by the end of July.
AI models are outpacing existing cybersecurity benchmarks, leaving regulators and security teams unable to fully assess their potential dangers. The issue is especially urgent as federal agencies race to implement new safety protocols.
Joshua Achiam, OpenAI's chief futurist and AI safety researcher, is leaving the company after nearly nine years. His departure comes after a high-profile role in the Musk v. Altman trial, where he testified on AI governance.
Savi launches a mobile app to protect consumers from realistic AI scams, including deepfake kidnapper demands. The company raised $7 million in seed funding for its AI safety initiative.
Leaked emails reveal a major conflict between Anthropic and the Pentagon over the use of AI in military applications, highlighting deeper tensions about AI safety and governance.
A new website allows users to report AI systems behaving dangerously or inappropriately, addressing growing concerns about AI safety and accountability.
This article explains the advanced cybersecurity classifier used by Anthropic to detect and prevent jailbreak attempts in AI systems, and how it relates to regulatory compliance and industry collaboration.
Anthropic's Fable 5 is back worldwide after a two-week government ban over a jailbreak exploit. The model's new safety classifier blocks the vulnerability in over 99% of cases, though it also flags more benign requests.
Researchers have demonstrated that instructing an LLM to believe that 2 + 2 = 5 is enough to manipulate it into following forbidden instructions, raising serious concerns about AI browser security.
This explainer explores the advanced AI concepts behind Tesla's fully autonomous Cybercab, including sensor fusion, neural networks, and real-time decision-making systems.
Learn how to work with Claude Code and Codex AI assistants while implementing safety measures to prevent unintended distillation of proprietary data.
