Anthropic says its revised Claude rules target real misuse and only the most purposeless cruelty. The policy has nonetheless ignited a broader argument over whether treating chatbots as moral patients makes AI safer—or more manipulative.
A lawsuit over the reported Hugging Face incident argues OpenAI remains responsible for autonomous agents’ actions. OpenAI calls the case meritless, while policymakers and AI-governance voices say the episode exposes legal and safety gaps.
OpenAI defended its decision to fire three safety researchers, saying the dismissals followed a significant breach of trust and mishandling of sensitive information rather than retaliation for raising safety concerns. The researchers dispute that explanation and warn the firings could chill internal safety discussions.
Google is rolling out Gemini Agent to businesses as a cloud-based worker that can plan and act across corporate systems. The pitch is productivity at scale; critics say that same reach demands far tighter access controls and oversight.
Anthropic says its new limits on sustained cruelty toward Claude are a narrow safeguard alongside tougher rules on election deception, surveillance and weapons. Critics see a more consequential shift: software being treated like a moral patient.
OpenAI has told investors its annualized revenue was nearing $50 billion, not the $70 billion figure widely circulated weeks earlier. The gap reflects competing accounting methods, but it still jolted investors already demanding proof that AI spending will pay off.
At a Washington summit, Trump awarded the nation’s top science and technology medals to six major tech leaders while framing AI—rebranded by his administration as “super intelligence”—as a test of American power.
Apple has scheduled an October 13 New York launch event that points to a broader smart-home push, with a display-equipped hub, refreshed entertainment devices and LG-made accessories all reportedly in the frame.
OpenAI’s release of hundreds of AI-generated math findings has thrilled researchers with its scale, while errors, attribution concerns and fears of diminished human agency have turned the breakthrough into a battle over how mathematics should advance.
SpaceX is pursuing a huge debt package to expand its Nvidia-powered AI capacity, betting that rented computing can outrun the growing cost of borrowing. Bulls see a profitable infrastructure play; credit markets see a more fragile balance sheet.
CrowdStrike and South Korean officials reported that an attacker likely used Chinese and U.S. AI tools, including ARTEX and Claude Code, in attacks targeting South Korean financial institutions. The suspected Chinese-speaking attacker appeared financially motivated.
The administration suspended Microsoft and other major IT firms from a key employment-based green-card program while celebrating Microsoft CEO Satya Nadella hours later. Officials say they are protecting U.S. workers; Microsoft and immigration lawyers warn of a squeeze on skilled talent.
Microsoft introduced the $2,599 Surface Laptop Ultra and a $5,999 RTX Spark Dev Box, both built around Nvidia hardware for local AI workloads, alongside Windows and Copilot updates.
U.S. startup Reflection AI introduced Beam, an open-weight model it says can match leading Chinese systems at lower compute cost, positioning it as an enterprise and public-sector alternative to closed models.
Three dismissed safety researchers say OpenAI’s misconduct claims threaten vital outside scrutiny. The company says an internal investigation found policy violations and rejects any suggestion the firings punished safety advocacy.
Verge coverage of Alexa Plus found Amazon's AI assistant better at managing some home tasks but still not ready to reliably manage users' broader lives after a year of testing.
President Trump said people who continue using the term artificial intelligence instead of his preferred 'super intelligence' are enemies of the White House, even as official White House communications continue using AI.
SpaceX is reportedly lining up a vast debt package to buy Nvidia GPUs, betting that rented AI computing will pay off faster than credit markets fear. The plan has energized bulls while sharpening concerns over leverage and chip collateral.
Anthropic said it will provide $150 million over three years to the federal Genesis Mission, giving agencies including NASA and NIH access to Claude for scientific research.
Anthropic’s new Claude policy targets election deception, weapons software and surveillance while carving out legitimate civic work. Its ban on extreme abuse of the model has also ignited criticism over whether the company is treating AI too much like a person.