AI & Enterprise

Claude trace left in U.S. lawmaker draft bill after claim of AI proofreading

A draft amendment from U.S. Republican Representative Anna Paulina Luna (안나 파울리나 루나) drew controversy after it contained the phrase “Claude responded:”, suggesting material may have been pasted from Anthropic’s Claude chatbot. Luna acknowledged on X that staff used AI while correcting draft wording but later said it was limited to spell and grammar checks of a summary, not the amendment text. Her posts were later deleted.

AI & Enterprise

OpenAI to limit early access to GPT-5.6 for some companies

OpenAI is delaying a full launch of its next flagship model, GPT-5.6, and will begin a limited release for selected corporate customers. The Trump administration requested a phased rollout over potential security concerns, with customer access expected to require case-by-case approval during the pilot period. The move follows tougher action previously directed at Anthropic. The approach has raised questions about consistency in U.S. AI controls and how future model releases may be affected.