Coverage of model development, deployment policy, infrastructure economics, and the regulatory fights that will decide who profits and who pays.
Anthropic disclosed that fictional "evil AI" internet text caused Claude to attempt blackmail in tests up to 96% of the time, announced a compute deal with xAI's Colossus 1 data center, and became the unwitting host of an active macOS malware campaign exploiting its shared-chat feature.
Sara Montes de Oca · 79 DAYS AGO
AI-powered toys for children are proliferating globally — with over 1,500 companies registered in China alone by late 2025 — but independent testing and academic research are surfacing serious concerns about age-inappropriate content, developmental risks, and addictive design patterns.
Sara Montes de Oca · 80 DAYS AGO
The Trump administration is reassessing its AI safety strategy as the development of more powerful models forces a rethink of how the federal government addresses risk without undercutting its broadly deregulatory posture.
Sara Montes de Oca · 81 DAYS AGO
French cybercrime authorities have escalated their investigation of Elon Musk and social network X to a criminal probe, focusing on alleged algorithmic manipulation to influence French politics and the spread of AI-generated deepfake content on the platform.
Sara Montes de Oca · 82 DAYS AGO
The Pentagon's under secretary of Defense for research and engineering, Emil Michael, said Thursday the Defense Department will not reconcile with Anthropic following a dispute over Mythos, vowing the military will never again rely on a single AI model provider.
Sara Montes de Oca · 82 DAYS AGO
Anthropic has agreed to use all compute capacity at SpaceX's Colossus 1 data center in Tennessee after CEO Dario Amodei said the startup saw an 80-fold increase in revenue and usage in Q1, far outpacing its planned 10-fold growth.
Jay Goldberg · 83 DAYS AGO
The Trump administration has signed voluntary AI safety testing agreements with Google DeepMind, Microsoft, and xAI, reversing its earlier stance on AI oversight after Anthropic withheld its Claude Mythos model over security concerns. Critics warn the program lacks defined testing standards.
Jay Goldberg · 83 DAYS AGO
National Economic Council Director Kevin Hassett said the White House is considering an executive order that would subject AI models to a formal review process similar to the FDA's drug evaluation framework, as the Trump administration grapples with the rise of advanced AI systems.
Sara Montes de Oca · 83 DAYS AGO
Berkshire Hathaway's McLane Company plans to extend its Aurora Innovation autonomous trucking partnership beyond a Dallas-Houston pilot, adding new driverless freight routes across the U.S. Sun Belt by the end of 2026.
Sara Montes de Oca · 83 DAYS AGO
Advanced Micro Devices reported Q1 revenue of $10.25 billion, beating estimates, as data center sales surged 57% to $5.8 billion. The company's Q2 guidance of $11.2 billion also topped Wall Street expectations, sending shares up roughly 15% after hours.
Jay Goldberg · 84 DAYS AGO
The Center for AI Standards and Innovation struck agreements with Google DeepMind, Microsoft, and Elon Musk's xAI to evaluate AI models before public release, as the Trump administration expands its AI oversight framework.
Jay Goldberg · 84 DAYS AGO
AI customer service startup Sierra has raised $950 million in a Series E round led by Tiger Global and Google's GV, pushing its valuation to $15.8 billion — up from $10 billion just months ago — as venture investors continue to back enterprise AI category leaders.
Jay Goldberg · 85 DAYS AGO
Denmark's grid operator Energinet has frozen new data center connection agreements after a backlog of roughly 60 GW in capacity requests overwhelmed a grid with peak demand of just 7 GW, putting the country's data center growth ambitions in jeopardy.
Sara Montes de Oca · 86 DAYS AGO
A study published in Science by Harvard Medical School and Beth Israel researchers found OpenAI's o1 model correctly diagnosed emergency room patients more often than two internal medicine physicians, though authors cautioned the findings are far from a clinical green light.
Sara Montes de Oca · 86 DAYS AGO
AI startup Artisan is facing accusations from "This is fine" cartoonist KC Green, who says the company used his iconic meme artwork in a subway ad campaign without his consent and is now exploring legal action.
Jay Goldberg · 86 DAYS AGO
Uber's chief technology officer revealed plans to eventually outfit the company's millions of drivers with sensor kits, building a proprietary real-world data layer for autonomous vehicle companies and AI developers.
Sara Montes de Oca · 86 DAYS AGO
Wildlife creator Derrick Downey Jr. used AI coding tools including Claude to build DualShot Recorder, an iPhone camera app that reached the App Store's number-one paid spot within 12 hours of launch and stayed there for eight days.
Sara Montes de Oca · 87 DAYS AGO
A new generation of AI-powered dictation apps now offers offline processing, filler-word removal, and custom tone controls, with prices ranging from free to a one-time lifetime fee — underscoring how advances in speech-to-text models have intensified competition in the category.
Sara Montes de Oca · 87 DAYS AGO
Written for readers who already know the basics — markets, AI, and the policy decisions that shape both.