EGI-BENCH Unifies Eight Enterprise AI Agent Benchmarks Under Single Framework
Researchers release EGI-BENCH, consolidating eight enterprise AI benchmarks with a new six-dimensional scoring system for evaluating LLM agents.
arxiv.org
The daily briefing
AI news. In perspective.
Latest publication
How stories are selectedIn focus
Anthropic Proposes AI Safety Bill With Thresholds for Blocking Model Releases
Anthropic has drafted legislation that would allow blocking AI models posing risks of 50+ deaths or $1B+ in damages, according to Politico.
Source: techcrunch.com
Also in the briefing
All newsMeta Pauses Muse Sharing Feature After Researchers Discover Public AI Chats
Meta suspended its Muse chat-sharing feature after security researchers found over 6,800 conversation links publicly accessible via DNS records.
theverge.com
Researchers propose Petri net framework to enforce tool-access rules for LLM agents
A preprint describes using Petri nets to model preconditions, approvals and budgets around tool calls, refusing requests—including injected ones—that lack required marks regardless of phrasing.
arxiv.org
OpenAI Acquires AI Biotech Startup Orchestra Bio
OpenAI acquired Orchestra Bio, a four-year-old startup that creates AI agents for drug discovery operations, to enhance its life sciences capabilities.
techcrunch.com
The longer view
Context from the archive
Historical coverage, revisited. A rotating selection refreshed with each day’s site build.
Historical coverage ·
Meta's Llama 3 Release Marked Open Source AI's Coming of Age
How Meta's April 2024 release of Llama 3 demonstrated that open-source models could match proprietary AI systems in performance and scale.
Revisit the storyHistorical coverage ·
Anthropic's $50 Billion Infrastructure Bet Marks Pivotal Moment in AI Scale Race
How Anthropic's massive November 2025 infrastructure commitment signaled a new phase in enterprise AI competition and domestic manufacturing.
Revisit the storyHistorical coverage ·
ChatGPT's Unprecedented Debut – The AI Moment That Captured the World
On November 30, 2022, OpenAI launched ChatGPT as a free research preview, swiftly reaching 1 million users and igniting global discourse on AI's potential impact.
Revisit the story