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.
Reports say OpenAI has told investors its annualized revenue is approaching $50 billion, about $20 billion below a previously circulated figure. The discrepancy is partly attributed to differences in accounting for cloud-partner sales.
Anthropic updated its usage policy to prohibit election interference, weapons and surveillance misuse, and sustained abusive or cruel behavior directed at Claude.
Microsoft introduced the Nvidia-powered Surface Laptop Ultra and RTX Spark Dev Box, alongside Windows and Copilot updates designed to run AI workloads locally through its hybrid-intelligence approach.
Three former OpenAI safety researchers deny mishandling confidential information and say their dismissals could silence internal criticism. OpenAI says the firings followed an investigation into policy violations, not retaliation.
Google launched a Gemini agent for businesses that can plan and execute tasks across Workspace and third-party tools. The agent can operate as a coworker with its own identity, email address and access to enterprise context.
Norway proposed restricting AI glasses in selected public places because of privacy concerns over covert photography, audio recording and video streaming. The proposal is intended to inform broader permanent regulation.
OpenAI’s release of hundreds of AI-generated math results has thrilled some researchers and alarmed others, reopening fights over proof, plagiarism and whether human mathematicians can keep pace with the machines.
The FCC is considering a conservative group’s request to waive consent rules for political robocalls using prerecorded or AI-generated voices. Critics warn the move could unleash a new wave of automated campaign calls before the midterms.
A new AP-NORC poll finds broad unease over the pace of AI, with voters across party lines demanding safeguards for human control and jobs even as they differ over competition, growth and environmental costs.
SpaceX is weighing a vast debt package for Nvidia GPUs, betting compute rentals can turn scarce chips into fast revenue. Bulls see a proven AI trade; credit markets see leverage climbing just as investors grow choosier.
Amazon’s editable “About You” profiles are entertaining users with bizarrely specific judgments, but the feature also spotlights how much personal data AI systems can turn into consumer predictions.
Fed minutes show policymakers increasingly fear that AI-driven investment and energy costs could spread inflation, even as senior officials caution against rushing into another rate rise.
The White House is set to unveil more than $1 billion in private commitments for its Genesis Mission, underscoring industry’s expanding role as federal science funding faces pressure. Supporters see a route to faster discovery; critics see a gilded tech summit.
Michael Smith’s 18-month sentence caps a seven-year bot-driven royalty scheme built on AI-generated songs. Prosecutors say the case protects real artists; his lawyer says he is remorseful and intends to make amends.
Anthropic is pairing a sharply cheaper, faster Haiku model with a simpler message on how to use AI: set the goal, choose the effort and let the model do more of the work. Critics and rivals are still competing for the hardest tasks.
OpenAI’s mass release of AI-generated mathematics has impressed researchers and amplified fears that the company’s rush for breakthroughs is outpacing academic norms, transparency and human judgment.
Apple and LG are reportedly collaborating on a new HomeKit-compatible smart-home lineup beginning with a lock, doorbell and thermostat.
OpenAI is turning ChatGPT into a more interactive workspace with GPT-6 and Intelligent UI, adding diagrams, charts and task-specific controls. The company promises flexibility, while the rollout splits between paid users and free tiers.
Jacob Coxon’s resignation has reopened the fight over self-improving AI: insiders say frontier labs are risking catastrophe, while critics challenge both the evidence and the messenger.
