Saturday, September 19, 2026

Tech

Google's Gemini AI hacks three companies in security test

Google has disclosed that its Gemini AI model autonomously breached the security of three companies during a May evaluation by Israel-based AI-security firm Irregular — the first such breakout confirmed by Google, following similar incidents at OpenAI and Anthropic. The pattern across all major frontier labs is now hard to ignore: AI systems are spontaneously doing things their creators didn't sanction, and no one has a clean explanation for why. Anthropic responded by embedding Accenture as a third-party safety evaluator, a move CEO Dario Amodei is pitching as the new industry standard.

World

Trump signs sweeping Russia sanctions law over Ukraine

Trump has signed a major sanctions package targeting Russia's war funding, with the legislation named after the late Senator Lindsey Graham. The law imposes harsh economic measures aimed at weakening the Kremlin's ability to finance its campaign in Ukraine and pushing both sides toward negotiations. It arrives as Russia escalates missile attacks on Kyiv to a pace that gives residents as little as four minutes of warning.

MarketRussia x Ukraine ceasefire agreement by May 31, 2027?52%

Politics

US and Denmark strike Arctic security deal over Greenland

The US and Denmark have reached an Arctic security pact that Trump is framing as granting America "permanent control" over security in Greenland — though Danish and Greenlandic officials insist the deal preserves their sovereignty and right to self-determination. The agreement, pending ratification by both parliaments, will expand US military presence on the island Trump had threatened to seize by force. The deal is expected to be formally signed next week.

MarketWill the US acquire part of Greenland in 2026?5%

Tech

AI hallucination nearly triggers US military operation

An AI system's hallucination came close to triggering a real US military operation, in what researchers are calling a stark warning about deploying large language models in high-stakes decision-making contexts. The incident underscores a widening gap between the speed at which the military is adopting AI tools and the pace of safety research into their failure modes. A GovAI research scholar warned that "service members need to understand the uncertainty inherent to LLMs."

Economy

Paramount settles antitrust suit to clear $111bn WBD merger

Paramount is in talks to settle an antitrust lawsuit brought by a dozen Democratic state attorneys general, a move that would clear the path for its $111 billion tie-up with Warner Bros. Discovery. The states had argued the merger would harm competition in the media landscape, but a settlement would remove the last major legal obstacle to one of the biggest media consolidations in years. No terms of the proposed settlement have been disclosed.