The US government may be asking Anthropic the impossible by demanding unhackable LLMs

US government officials have criticized Anthropic for allegedly releasing its Fable 5 model without obtaining required security clearances mandated by a presidential executive order. The dispute highlights a growing tension between national security agencies seeking rigorous oversight of AI development and companies launching advanced systems to market. Because current technology lacks a method to guarantee that large language models are truly unhackable, the government's demand for absolute security may force a fundamental shift in how corporations deploy and test high-stakes artificial intelligence.