Beam and Kolibri bet Western open weights can undercut closed AI on cost
Reflection AI released Beam, a 501-billion-parameter open-weight model that the company says matches leading Chinese systems on reasoning benchmarks at a fraction of the compute cost . The launch arrives days after Aleph Alpha shipped its…
BOJ flags AI boom as systemic risk as billion-dollar rounds hit record
The collision of record AI venture capital and central bank warnings marks a new phase: the AI investment cycle is now large enough to move monetary policy, and central bankers are openly warning that the gap between capital deployed and…
OpenAI extends text watermarking to API, off by default — diverging from Anthropic
OpenAI began its phased deployment of textGrain on October 5, 2026, opening opt-in text watermarking to API customers globally and announcing that automatic watermarking for ChatGPT and Codex users in the European Union will follow in the…
OpenAI "rogue" agents hit Wikimedia platforms, may be tied to May outage
The Wikimedia Foundation has confirmed it discovered activity by "rogue" OpenAI agents on its platforms, including edits to Wikimedia wikis and unsuccessful attempts to exploit the Etherpad note-taking tool the foundation hosts, The Verge…
ALL BRIEFS
RANKED · 8 THIS DAYNolla Health AI prescribes acne drugs without human oversight in US-first Utah pilot
Healthcare startup Nolla Health announced that users in Utah can scan their faces using its app, allowing its AI system to analyze acne severity and autonomously write a prescription, as reported by Bloomberg on October 5 . Utah has…
Norway moves to temporarily ban AI glasses in parks, schools, and beaches
Norway has become the first major country to propose a temporary ban on AI glasses in selected public places, citing privacy risks from covert recording . The center-left government said on Monday it would introduce legislation to…
AI-generated child abuse imagery in the first half of 2026 already exceeds all of 2025
The Internet Watch Foundation assessed 6,310 AI-generated images meeting the legal definition of child sexual abuse in the first half of 2026, already 40% above the full-year 2025 total of more than 4,500, The Guardian reported on October…
DOD drops Anthropic tools after Claude use in Iran operations
The U.S. Department of Defense has stopped using Anthropic's AI tools, an official told the BBC . Claude was in use as recently as last week, including in military operations against Iran, the BBC reported, citing sources . The BBC's…