tech
100 cyber experts say Fable 5 ban hurts defenders
Three days after the US government ordered Anthropic to shut down Fable 5 and Mythos 5, roughly 100 of the world’s most prominent cybersecurity professionals have published an open letter demanding the ban be reversed. Their argument is blunt: pulling the best AI tools from defenders while adversaries keep building is not safety, it is sabotage.
TL;DR
- Approximately 100 cybersecurity professionals have signed an open letter calling for the reversal of the US government's order to shut down Anthropic's Fable 5 and Mythos 5.
- The professionals argue that removing advanced AI tools from defenders while adversaries continue development constitutes sabotage and does not enhance safety.
- They state that the identified exploit is not unique and other leading AI models can reveal similar vulnerabilities without being banned.
- The ban has created market uncertainty and risks America's AI leadership without sufficient justification.
- Amazon, Anthropic's largest investor and cloud host, discovered the vulnerability and escalated the findings to high-level government officials, raising questions about commercial rivalry.
- The shutdown of Fable 5 coincided with the launch of China's Zhipu AI GLM-5.2 model, which cited the ban as evidence of the unreliability of US AI models.
- Cybersecurity professionals utilize frontier AI models for vulnerability hunting, detection rule generation, and malware analysis.
- The letter emphasizes that adversaries can access open-source alternatives or foreign models, while defenders are left with less capable tools.
- Prediction markets suggest the ban on Fable 5 is likely to be temporary, with high odds of its return before July 1st.
- The European Union and India are also increasing pressure and accelerating their own AI ambitions following this episode.