Summary based on reports from Axios and Yahoo News.
OpenAI’s new advertising pilot has reportedly generated about $100 million in annual recurring revenue within two months. The company is projecting $2.5 billion in ad revenue in 2026 and as much as $100 billion by 2030. This indicates that advertising is becoming a core part of OpenAI’s business model and raises concerns about whether chatbots will primarily serve users or advertisers.
OpenAI is developing a new cybersecurity-focused product built on its advanced models. This move comes as rivals like Anthropic have begun restricting access to powerful systems, such as the Mythos Preview model, because of fears they could enable sophisticated hacking and even autonomous attacks on critical infrastructure if misused.
OpenAI CEO Sam Altman has released a detailed policy blueprint arguing that looming AI “superintelligence” requires a social contract on the scale of the New Deal. His proposal includes new approaches to taxation, regulation, and redistribution of AI-driven wealth, along with broad, low-cost access to AI tools so workers and communities are not left behind.
Meta is signaling that it will keep open-sourcing versions of its next-generation AI models. The company is positioning itself as a democratizing force for access to cutting-edge AI and as a U.S.-based open alternative for developers. This stance comes as OpenAI and Anthropic indicate that their own upcoming models will be major advances but will remain closed.
OpenAI chief scientist Jakub Pachocki said in a recent podcast that AI systems are approaching the capability level of a “human research intern.” He cited breakthroughs in coding and early progress in math and physics research, and suggested that AI will increasingly handle complex, multi-step technical work with less human oversight.