Microsoft has announced the termination of its exclusive partnership and revenue-sharing agreement with OpenAI, with both companies agreeing on a new stage for their future partnership.
A developer reported that an OSS AI agent they built achieved a score of 65.2% on the TerminalBench benchmark, surpassing Google's official score.
World has announced partnerships with Tinder and Zoom to provide identity verification services using iris scanning and facial recognition technology.
Posts on X (formerly Twitter) suggest that Claude Code features provided by Anthropic may no longer be included in the Pro plan.
OpenAI has released ChatGPT Images 2.0, a new version of its image generation capabilities.
As computational functionalism dominates discussions on AI consciousness, the "Abstraction Fallacy" has been proposed, arguing that this view misinterprets the relationship between physics and information.
A case has been reported where the misuse of an API key in a project with Firebase AI Logic enabled resulted in charges exceeding 50,000 euros within 13 hours.
Google has released a native version of the Gemini app for macOS. Users can invoke it via Option+Space to receive immediate assistance through screen sharing.
On April 15, 2026, Google announced Gemini Robotics-ER 1.6, a new model designed to enhance the physical reasoning capabilities of robots. With improved spatial understanding and instrument reading functions, it is available via the Gemini API starting today.
NBC News has revealed via an obtained letter that Apple warned xAI in January that Grok could be removed from the App Store for failing to address the generation of sexual deepfakes.
The San Francisco residence of OpenAI CEO Sam Altman was targeted by gunfire in the early hours of Sunday, following an incident on Friday where a Molotov cocktail was thrown. Police announced the arrest of multiple suspects involved in the two incidents.
MiniMax has open-sourced its AI model M2.7. Through an autonomous improvement cycle of self-analysis and code correction, performance has increased by 30%, and the model is now available on HuggingFace.
OpenAI's Sam Altman has shared his views on the importance of AI democratization and safety measures following an incident where a Molotov cocktail was thrown at his residence.
Sam Altman commented on the incident where a Molotov cocktail was thrown at his home, stating that AI is the most powerful tool for expanding human capabilities and that ensuring its democratization and safety is essential.
On April 10, 2026, a Molotov cocktail was thrown at the home of OpenAI CEO Sam Altman. Multiple news organizations, including the BBC and CNN, report that a suspect has been arrested.
A 20-year-old man has been arrested for making threats at OpenAI's San Francisco headquarters and throwing a Molotov cocktail at the home of CEO Sam Altman.
OpenAI has expressed support for Illinois bill SB 3444, which would exempt AI developers from liability under certain conditions, even if their models cause significant societal harm.
OpenAI has expressed support for Illinois bill SB 3444, which would exempt advanced AI developers from liability for catastrophic harm caused without intent or gross negligence. AI policy experts point to this as a turning point in OpenAI's legislative strategy.
According to a report by The New Yorker, OpenAI CEO Sam Altman has limited experience in programming and machine learning, and occasionally confuses basic AI terminology.
Government ethics records show that Emil Michael, the Under Secretary of Defense for Research and Engineering, sold xAI shares worth between $5 million and $25 million—realizing gains of up to 4,800%—while the Department of Defense held two contracts with the company.