Latest articles in Technology
OpenAI has deferred the public rollout of its next-generation AI model, GPT-5.6, following a request from the White House to provide U.S. government agencies with early access to frontier AI systems. The company expressed concerns that such restrictions should not become the norm, raising questions about the balance between national security and open AI development.
The US government has partially reversed its suspension of Anthropic's Claude Mythos 5 AI model, allowing release to 'trusted' American organizations for controlled testing. The decision comes as OpenAI defers public rollout of GPT-5.6 following US requests for early access to frontier models. Samsung announced a record $648 billion investment plan in South Korea over the next two decades, driven by AI chip demand.
OpenAI announced it will delay the public release of its next-generation GPT-5.6 model following a request from the Trump administration for early access to evaluate frontier AI safety. The US government is seeking preferential access to advanced AI systems before broad deployment, citing national security concerns. The delay marks an escalation in government oversight of the AI industry and has sparked debate about the balance between innovation and regulation.
Oracle reduced its workforce by 21,000 employees in fiscal 2026, a 13% drop, as the cloud computing giant accelerates its shift toward AI-driven automation. The company confirmed that AI systems have replaced roles in customer support, database management, and software testing. Oracle now employs around 141,000 full-time workers, down from 162,000 a year ago. The layoffs reflect a broader trend across the tech industry as companies invest heavily in AI.
OpenAI announced its first custom-designed artificial intelligence chip, built in collaboration with Broadcom, marking a major step toward reducing the company's reliance on Nvidia's graphics processing units. The new chip is designed specifically for inference workloads, the process of running trained AI models to generate responses. The move reflects a broader industry trend of AI companies designing their own silicon to optimize performance and cost for their specific needs.
Anthropic launched Claude Tag, an AI assistant that integrates with Slack and other workplace tools to learn company-specific context and terminology. The tool indexes internal conversations and documents to provide more relevant responses, but privacy advocates warn it could enable unprecedented workplace surveillance. The launch comes amid ongoing tension between Anthropic and the Trump administration over AI regulation.
Norway announced a near-total ban on generative AI tools in elementary schools starting in late August, Prime Minister Jonas Gahr Støre said Friday. Children aged 6 to 13 will generally not be allowed to use AI in school, while older students face supervised age-tiered access. The government cited concerns that AI erodes children's ability to develop cognitive and social skills, and that the technology does too much of the work children need to learn themselves. Norway, one of the first countries to ban smartphones in schools, is also considering new legislation to promote book use in classrooms.
White House talks with AI company Anthropic have shifted to setting AI security rules, Politico reported. Meta executive Joel Kaplan detailed the company's AI, data center and workforce plans at POLITICO's Energy Summit. The Federal Energy Regulatory Commission chair discussed the agency's AI agenda, as fast-tracked power plants fuel AI's booming energy demands.
The US government blocked Anthropic from releasing its Fable 5 AI model, citing national security concerns, in what experts call the most aggressive federal action yet against a private AI company. The ban has sparked debate about export controls, with Anthropic's CEO clashing with the White House over the restrictions. Separately, Amazon announced plans to sell its own AI training chips directly to customers, challenging Nvidia's dominance. ChatGPT's market share slipped below 50% for the first time as competitors gain ground. AI data centers also secured a government-mandated fast lane to the electrical grid.
Amazon announced a $50 billion investment in OpenAI as part of a broader strategic partnership that will see OpenAI run more workloads on AWS infrastructure, including a commitment to deploy 2 gigawatts of Amazon's Trainium AI chips. The deal is part of a $110 billion funding round for OpenAI that values the company at $730 billion. Analysts estimate the $100 billion cloud services component could add roughly $17 billion a year in AWS revenue—about 11% of expected 2026 revenue. The partnership reshapes the competitive AI landscape.