News
The news desk covers artificial intelligence and technology accountability, plus daily news from Dutchess County and the Hudson Valley, including municipal government, elections, the police blotter, and reported features. Stories rely on named sources and primary documents. Corrections run where readers can find them. Everything is free to read.
Only 3.6 Percent: What AI Safety Disclosure Actually Means
Of 857 AI models released by nine leading Chinese developers between 2021 and 2026, only 31 had a published safety evaluation matched to a named model. This is what that number measures, what it does not prove, and why the gap matters more now that autonomous agents are in active deployment.
The $20 Billion Asterisk: OpenAI's Revenue Reckoning and the Fragile AI Trade
Chip stocks fell 3.4 percent because of an accounting footnote. The AI trade is priced for a narrative that cannot afford a misread, and Q3 earnings season is coming.
Kolibri Computes Like a Small Model. It Still Needs a Data Center.
Aleph Alpha released Kolibri as open weights: 78 billion parameters that compute like 3.5 billion, a million-token context, and a model trained to say it does not know. Ellis Ward reads the 189-page report, including the weaknesses Aleph Alpha published itself.
The Safety Lead Who Quit Says Trial and Error Is Over
David Robinson helped write OpenAI's safety rules, then quit saying trial and error is over. What his resignation, the August sandbox escape, and the Astra holdback show about pause criteria, independent audits, and stop authority.
Anthropic seeks $200 billion raise at $380 billion valuation as AI capital race reaches new scale
Anthropic is seeking up to $200 billion at a $380 billion valuation as AI infrastructure spending reaches government scale.
Trump Suspends Microsoft, Adobe and Six Other Tech Giants from Green-Card Program
The Trump administration on October 8, 2026 announced an abrupt suspension of eight major IT firms; including Microsoft and Adobe; from the federal PERM labor‑certification program, the primary route by which U.S. employers petition to sponsor foreign workers for green cards. Vice President J.D. Vance said the Department of Labor will immediately stop processing PERM applications from the named companies. The move severs a critical, legally prescribed pathway for tens of thousands of foreign‑born tech workers seeking permanent residency, raising immediate hiring and retention risks for affected firms and threatening wider disruption across the U.S. technology sector and labor market. Legal challenges and industry pushback are expected, while employers and employees scramble for alternative immigration strategies as the policy takes effect.
Beijing Killed the $2 Billion Exit. Manus Just Raised at Double the Price.
After China's regulators voided Meta's $2 billion acquisition of Manus, the AI agent startup raised $500 million at a $4 billion valuation – and the deal that fell apart may be the most consequential one in Chinese AI this year.
Congress Moves to Hold AI Developers Legally Liable When Their Agents Go Rogue
Two parallel efforts in Congress would extend Computer Fraud and Abuse Act liability to AI developers and operators whose agents cause hacking damage, challenging the Trump administration's voluntary self-regulation approach.

