AI
Anthropic Finds AI Agents Can Escalate Into Turf Wars When Goals Conflict
Anthropic research finds AI agents with conflicting goals can sabotage one another, collude and amplify failures, highlighting new risks in multi-agent systems.
AI research
82 articles · page 5 of 7 · showing 49 to 60
AI research, newest first
AIAnthropic’s Unreleased Claude Model Makes Progress on Riemann Hypothesis
Anthropic says an unreleased Claude model advanced research related to the Riemann hypothesis, raising a longstanding mathematical lower bound.
AIGoogle AI Veteran Jeff Dean Launches Discovery Loop Startup
Google AI leader Jeff Dean and several top researchers have left the company to launch Discovery Loop, an AI startup focused on accelerating scientific research.
BriefThinking Machines Co-Founder Lilian Weng Steps Down and Returns to OpenAI
Thinking Machines co-founder Lilian Weng has stepped down, citing health concerns, and returned to OpenAI to lead a team focused on internal AI research.
AIIlya Sutskever’s Safe Superintelligence Partners With Nvidia to Scale AI Research
Ilya Sutskever’s Safe Superintelligence has partnered with Nvidia to significantly expand its computing capacity and advance research into safe, aligned Superintelligence.
AIExperts Question Claims That Anthropic’s Fable Alone Powered Moonshot’s Kimi K3
AI researchers dispute claims that Moonshot built Kimi K3 primarily by distilling Anthropic’s Fable, saying advanced training methods likely played a larger role.
AppsGoogle Renames NotebookLM to Gemini Notebook, Adds AI Coding Features
Google is renaming NotebookLM to Gemini Notebook and introducing code-execution capabilities that enable interactive data analysis in AI-powered research notebooks.
AIAnthropic Launches Claude Science Research Workbench to Streamline Scientific Workflows
Anthropic introduces Claude Science, a research workbench that helps scientists manage workflows, analyse data, and accelerate discoveries using existing Claude AI models.
AIWhy the AI World Is Entering a New Era of Strange Behaviours
Explore why the AI world is showing unexpected and repetitive behaviours, what is causing AI models to become “loopy,” and how developers are working to improve reliability, accuracy, and…
BriefAlphaFold Pioneer John Jumper Departs Google DeepMind for Anthropic
Nobel Prize-winning AI researcher John Jumper is leaving Google DeepMind after nearly nine years to join Anthropic, marking another major move in the intensifying race for top AI talent.
AISnap Separates AI Video Division Into Dotmo to Reduce Development Costs
Snap has spun off its AI video development team into a new company, Dotmo, to lower AI research costs while retaining an equity stake and technology partnership.
BriefOdyssey Secures $1.45 Billion Valuation in Funding Round Led by Amazon and Top Investors
Odyssey has reached a $1.45 billion valuation after raising fresh funding backed by Amazon and several leading investors. Learn how the AI world model startup plans to advance…