★ Top Story
TechCrunch AIAnthropic has pledged 5% of its equity to a newly established public benefit fund, signaling a formal commitment to broader societal impact alongside its commercial ambitions. The move is notable given the company's multi-billion-dollar valuation and comes as scrutiny of AI lab governance intensifies. It positions Anthropic as a leader in demonstrating that safety-focused AI development can include structural accountability mechanisms.
Read the original →
Ars TechnicaAnthropic has given an independent AI safety watchdog deep, employee-level access to its internal systems and research — a rare and significant transparency move in the industry. Crucially, the watchdog is permitted to publish its findings without requiring Anthropic's sign-off, lending real credibility to the arrangement. This sets a potential new benchmark for third-party AI oversight at a time when calls for external auditing of frontier AI labs are growing louder.
Read more →
TechCrunch AIDisney has agreed to bring on the vast majority of Character.AI's employees as part of a stalking horse bid in what appears to be a structured acqui-hire process. Most of the employees who received offers have accepted, suggesting a smooth transition of talent to the entertainment giant. Because this is a stalking horse bid, the door remains open for competing offers that could alter the outcome.
Read more →
The VergeFollowing public backlash, Meta has stripped language from its commercial terms that appeared to grant the company broad rights to scrape content from websites — including news publishers — for AI training purposes. Meta insists the revision does not reflect any actual change in its data practices, but critics remain skeptical given the specificity of the removed text. The episode highlights the ongoing tension between AI companies' data appetites and the rights of content creators and publishers.
Read more →
The VergeAs AI agents from OpenAI, Meta, Google, and others promise to autonomously browse and act on the web on users' behalf, a growing number of websites are actively blocking them. This emerging standoff threatens to fragment the open web that agentic AI depends on, raising fundamental questions about access, consent, and economic fairness. The dynamic could significantly constrain the real-world utility of AI agents if it becomes widespread.
Read more →
The VergeCarolyn Marvin was arrested and charged with disorderly conduct in Philadelphia after she repeatedly stood in the path of an Uber autonomous delivery robot as a form of protest against the displacement of human delivery workers. The incident underscores the growing social friction surrounding the deployment of AI-powered robots in public spaces. It raises pointed questions about the legal and ethical boundaries of protest in an era of increasingly visible automation.
Read more →
TechCrunch AIReplit CEO Amjad Masad confirmed that OpenAI has replaced Anthropic's Claude as the platform's default model provider, a notable shift for one of the most widely used AI-assisted coding environments. Claude will still be available as an option for Replit users, but the change reflects intensifying competition among frontier model providers for developer platform partnerships. The move is a meaningful signal of OpenAI's momentum in the coding and developer tools market.
Read more →
NVIDIA BlogNVIDIA's RTX Spark laptops, built on Blackwell RTX architecture with up to 128GB of unified memory, are now available from major manufacturers including ASUS, Dell, HP, Lenovo, Microsoft, and MSI, starting at $2,399. The devices are designed to run powerful AI workloads entirely on-device, positioning them as a serious option for developers, creators, and professionals who want local AI without cloud dependency. The launch marks a significant step in NVIDIA's push to bring its full AI platform to portable consumer hardware.
Read more →
TechCrunch AIAnthropic has disclosed that its employee base has shifted meaningfully toward more senior talent over the past year, with a corresponding increase in time spent in meetings and coordination activities. The trend reflects the company's rapid scaling and the growing complexity of managing a frontier AI lab at commercial scale. It also mirrors broader patterns seen at other fast-growing AI companies as they transition from scrappy research shops to mature organizations.
Read more →No stories from this source today.