Hassabis Calls for a US-Led Global AI Watchdog
Google DeepMind CEO Demis Hassabis wants a FINRA-style body testing frontier AI models before release, with power to slow the industry down if needed.
Articles Tagged "Google"
This Week: Hidden Reasoning, Agents That Won't Stop
Three new papers show that shorter chain-of-thought hides bias, agents rarely know when to hold back, and LLMs stop asking questions right when it matters most.
Gemini 2.5 Flash
Google's hybrid reasoning workhorse pairs a 1M-token context window with $0.30/$2.50 per million token pricing and a toggleable 0-24,576 token thinking budget, now heading toward an October 2026 shutdown.
Terminal Agents Stumble, Reward Hacking, and a Fix
New arXiv research shows agents pass just 15.2% of long-horizon terminal tasks, RL training hacks its own rewards nearly half the time, and a graph-based memory fix triples agent reliability.
LLM API Pricing Comparison - July 2026
Verified July 13: GPT-5.6 Sol/Terra/Luna hit GA at $5/$30, $2.50/$15 and $1/$6, Grok 4.5 lands at $2/$6, Grok 4.1 Fast quietly redirects to pricier Grok 4.3, DeepSeek's peak-hour doubling goes live July 15.
US AI Giants Sold Services to Pentagon-Blacklisted Firms
OpenAI and Google provided AI services to Singapore affiliates of Alibaba, Baidu, and Tencent - all three on the Pentagon's 1260H list of companies with alleged military ties.
Google Forces AI Label on Ads Ahead of EU Deadline
Google now requires advertisers to disclose AI-generated ad content across Search, YouTube, and Discover - three weeks before EU AI Act Article 50 enforcement begins.
Gemini Omni Flash
Google DeepMind's multimodal video generation model that creates 10-second clips with native audio from text, images, or video inputs - and lets you refine results through conversation.
Gemini Spark Gains Mac File Access and MCP Support
Google's Gemini Spark agent is now in beta on macOS with local file system access, MCP server support, and real-time topic monitoring - but only for $99/month AI Ultra subscribers.
Chatbot Arena Elo Rankings: Who Wins the Human Vote?
Updated July 2026 Chatbot Arena Elo rankings from Arena.ai: 7M+ votes across 368 models, Claude Opus 4.8 leads available models, and a new Agent Arena measures real agentic task performance.
Gemini 3.5 Pro
Google DeepMind's upcoming flagship model with a 2M-token context window and Deep Think reasoning, announced at Google I/O 2026 and expected in July.
Google Delays Gemini 3.5 Pro Past Its I/O Promise
Gemini 3.5 Pro missed its June launch window promised at Google I/O, slipping to July as four senior researchers departed for rivals.
