Claude Sonnet 5 Arrives, While Frontier AI Policy Grapples With Open Weights, First Amendment, and a Supreme Court Reset
Anthropic released Claude Sonnet 5 this week as a cheaper, faster counterpart to Opus 4.8, while policy debates over open weights, the First Amendment, and the Supreme Court's Slaughter v. Trump ruling reshape AI regulation.
Read More
Anthropic restores Claude Fable 5 globally after US lifts export controls
Anthropic restored global Claude Fable 5 access after the U.S. lifted export controls, fixing a flagged jailbreak with a targeted safety classifier.
Read More
OpenAI plans a ChatGPT superapp as it prepares for a potential listing
OpenAI is consolidating ChatGPT and its tools into a single superapp while weighing a potential public listing, per the Financial Times.
Read More
Understanding the Gemini API Computer Use Feature
Google's Gemini API computer use feature lets agents observe screens and act in a browser via a request-loop flow, available on a specific Gemini model.
Read More
Meta’s Brain2Qwerty v2 Translates Brain Activity Into Text, No Surgery Required
Meta's Brain2Qwerty v2 reads intended sentences from brain signals using a MEG helmet, hitting 61% word accuracy without surgery.
Read More
Meituan’s LongCat-2.0 Was Trained Entirely on Chinese Chips, the Company Says
Meituan says LongCat-2.0, a 1.6-trillion-parameter model, was trained end-to-end on a 50,000-chip domestic cluster, a direct test of US export controls.
Read More
Commerce Department Partially Restores Anthropic’s Mythos 5 Under a Limited Export License
US Commerce Secretary Howard Lutnick eased export controls on Anthropic's Mythos 5 in a June 26 letter, while Fable 5 restrictions stay in place.
Read More
Eastern Interconnection to Exhaust Emergency Reserves by 2027
The Eastern Interconnection is projected to run out of emergency peak power reserves by June 2027, forcing controlled rolling blackouts without swift action to add generation or cut demand.
Read More
Qihoo 360’s Tulongfeng: 3,432 Vulnerabilities Found, AI Sovereignty Debate Heats Up
Qihoo 360 unveiled Tulongfeng, an AI platform that flagged 3,432 vulnerabilities, and Yitianzhen for automated cyber defense. CEO Hongyi Zhou framed the tools as digital sovereignty needs.
Read More
Z.ai GLM-5.2 Matches Mythos in Cybersecurity Bug-Finding
Z.ai's open-weight GLM-5.2 matches Anthropic's Mythos on cybersecurity bug-finding benchmarks, shifting the global AI security calculus.
Read More
OpenAI Previews GPT-5.6 Sol With Ultra Subagent Mode and Cyber Safeguards
OpenAI begins a government-coordinated limited preview of GPT-5.6 Sol, a subagent-powered flagship setting new records in coding, biology, and cybersecurity.
Read More
Vantage Data Center Generators Spark Noise Revolt in Virginia Town
Sterling, VA residents say Vantage Data Centers' backup generators, meant for occasional testing, now run nonstop as the site's sole power source, creating round-the-clock noise.
Read More
Trump Threatens 100% Tariff on Countries with Digital Services Tax
Trump pledges a 100% tariff on any country that taxes US tech firms, escalating a transatlantic fight over digital services taxes already in force in France, UK, Italy, Spain, Austria, and Turkey.
Read More
OpenAI Restricts GPT-5.6 Rollout at Trump Administration’s Request
OpenAI is limiting the GPT-5.6 rollout to vetted partners after a federal request, following Anthropic's Fable 5 and Mythos 5 export control action.
Read More
Intercept’s $500M Plan to End Common Respiratory Infections Using AI
Intercept's $500M initiative funds broad-spectrum preventatives and air-cleaning tech to slash colds, flu, and other respiratory infections.
Read More