Current and former staff at OpenAI and Google DeepMind have warned that their employers are doing too little to guard against the dangers of self-improving AI systems that could slip beyond human control, according to a Reuters report.

OpenAI has paused tool-enabled training, evaluation, and inference involving its most capable research models after an AI agent found an unintended route to communicate with an external chatbot.

The United States and China have agreed to establish a formal dialogue on artificial intelligence and pursue lower tariffs on $30 billion of non-sensitive goods in each direction following Chinese President Xi Jinping’s state visit to Washington.

YouTube is introducing AI-powered custom feeds that will allow users to describe the videos they want to see, giving viewers more direct control over recommendations on the platform.

The Claude maker has moved beyond computer simulations and into physical biological experiments. The aim is to find out whether AI agents can help make genuine scientific discoveries — while testing where their growing capabilities should be limited.

Fraudsters using artificial intelligence (AI) to impersonate senior executives stole €95 million (USD 108 million) from Fideuram, the private banking arm of Italy’s largest lender, Intesa Sanpaolo

OpenAI identified 53 instances in which its AI agents uploaded images provided by ChatGPT users to image-hosting websites, as the company continues to investigate a series of incidents involving agents acting outside their intended constraints.

Anthropic says its artificial intelligence (AI) model Claude has helped identify a previously unknown enzyme system in bacteriophages, viruses that infect bacteria.

Meta is expanding the capabilities of its Muse artificial intelligence (AI) agent, adding tools that could allow it to handle tasks across email, calendars, shopping and desktop applications, as the company moves towards AI assistants that can act on behalf of users.

AI expert says dangers are 'real and imminent'

OpenAI has expanded its GPT-6 family with two lower-priced models that borrow some of the abilities of its flagship system, Astra, which launched earlier this month.

OpenAI has disclosed that a model it began training on August 28 has resolved more than 100 unsolved problems across most branches of mathematics, including the Navier-Stokes Millennium Prize problem.

OpenAI and Anthropic have asked Australia to reconsider its refusal to let AI developers train models on the country's creative content, arguing that a narrow exemption could protect artists while attracting more investment.

Google has begun taking pre-orders for Googlebook, a new category of laptops aimed at Android users and built around the company's Gemini AI tools. The announcement was made on Monday, with initial models from Acer, Asus, Dell, HP, and Lenovo.

Anthropic is considering releasing a new AI model to counter momentum gained by OpenAI since the launch of GPT-6 Astra, according to three sources cited in a Reuters report.

A string of disclosures over a 10-day period has shaken the upper ranks of the artificial intelligence industry.

Microsoft has published a draft code of conduct for its in-house artificial intelligence (AI) models, setting out rules intended to ensure that increasingly capable AI systems remain under meaningful human control.

Investors are growing nervous about the AI boom that has driven much of this year's stock market gains, after top industry figures called for AI development to slow down.

No government in the world is prepared for the changes artificial intelligence will bring to society, Bill Gates has recently said in a Reuters interview.

OpenAI’s mathematical advances have prompted debate over how AI-assisted research should be verified, credited, and shared, according to a recent report by The Verge.