AI News Roundup – May 09, 2025
Trump AI chip policy to scrap Biden’s export controls
The Trump administration plans to replace Biden’s multi-tiered AI chip export controls, deemed too complex, with a potential global licensing system, while maintaining current export restrictions in the interim
OpenAI’s $3B Windsurf move: the real reason behind its enterprise AI code push
OpenAI’s reported $3B acquisition of AI-native IDE Windsurf aims to bolster its AI coding capabilities and secure dominance in the agentic AI future against competitors like Google and Anthropic
US AI execs give Congress policy wishlist for beating China
US AI executives from companies like OpenAI, Microsoft, and AMD urged Congress to implement policies such as increased infrastructure investment, streamlined regulations, and talent attraction to maintain a competitive edge over China in artificial intelligence
OpenAI will become a Public Benefit Corporation – here’s what that means
OpenAI’s for-profit arm is becoming a Public Benefit Corporation (PBC) to balance profit with its mission of ensuring AI benefits all, with its nonprofit arm remaining the largest shareholder
Introducing web search on the Anthropic API
Anthropic launched web search on its API, enabling Claude models to access real-time information and provide cited, up-to-date responses for developers building AI applications and agents
Introducing OpenAI for Countries
OpenAI has launched ‘OpenAI for Countries,’ an initiative to partner with nations to develop democratic AI infrastructure, including in-country data centers and customized ChatGPT for public services, while fostering national AI ecosystems
Will AI Really Be the Demise of Google Search?
While AI presents challenges to Google’s search dominance and ad model, with competitors emerging and antitrust concerns looming, Alphabet’s integration of AI into its services and existing advantages offer a potential buffer
Famous authors take on Meta in a first big test of AI and copyright law
Famous authors are suing Meta over AI and copyright infringement, raising questions about AI’s impact on creative work markets
ServiceNow and Nvidia’s new reasoning AI model raises the bar for enterprise AI agents
ServiceNow and Nvidia unveiled Apriel Nemotron 15B, an open-source reasoning LLM designed to enhance enterprise AI agents with improved latency, lower costs, and advanced capabilities
The AI Industry Has a Huge Problem: the Smarter Its AI Gets, the More It’s Hallucinating
Recent observations suggest a troubling trend where more powerful AI models are increasingly prone to ‘hallucinations,’ or generating false information, posing a major challenge for AI reliability

