AI Model Fable 5 Restored Following Emergency Suspension and Safety Upgrade
After a 19‑day emergency suspension by the U.S. government, Anthropic’s AI model Fable 5 is back online worldwide with enhanced safety controls and updated access policies.
Alan Turing AI Library
Browse the article collection, find matching topics, and open the full reading view in the library reading view. This page supplements the article experience with a searchable entry point.
After a 19‑day emergency suspension by the U.S. government, Anthropic’s AI model Fable 5 is back online worldwide with enhanced safety controls and updated access policies.
Five leading artificial intelligence laboratories have agreed to support a shared framework for evaluating jailbreak severity across models, aiming to bolster transparency and safety ahead of a targeted implementation date.
OpenAI’s upcoming model GPT‑5.6 is being released in a limited preview at the request of the U.S. government, marking a shift toward federal oversight of frontier artificial intelligence.
The U.S. federal government has taken significant steps in the past 24 to 72 hours to restrict early access to leading-edge AI models, marking a new level of regulatory oversight into AI deployment.
Intelligence agencies from the Five Eyes alliance caution that advanced AI models capable of enabling destructive cyber attacks are just months away, urging immediate defensive preparation.
The U.S. government has asked OpenAI to initially limit availability of its GPT‑5.6 model to approved partners. This move marks an intensification of regulatory oversight over frontier AI.
Anthropic’s AI model Mythos identified security weaknesses in classified U.S. government systems in just hours during a testing initiative—a development with significant implications for AI’s role in cyber‑defence and governance.
OpenAI has upgraded its specialized cybersecurity model GPT‑5.5‑Cyber, making it more capable and permissive for authorized defenders, while expanding international partnerships to bolster AI‑driven security.