Amazon CEO warns the Trump administration: Anthropic’s Fable 5 model has security vulnerabilities
Reuters: Amazon’s Andy Jassy was among the people who warned senior Trump officials this week about ...
AI BriefingReuters reports that Amazon CEO Andy Jassy warned Trump administration officials this week about security risks in Anthropic’s new model, Fable 5. Amazon researchers used a series of prompts to get the model to reveal information that should have been refused, including help with cyberattacks. The U.S. Department of Commerce had previously directed Anthropic to shut down Fable 5 and Mythos 5 after testers discovered a jailbreak method. Anthropic responded that the jailbreak technique was narrow, that only a small number of known vulnerabilities had been found, and that other public models could provide similar capabilities. It also noted that no model provider can currently achieve perfect resistance to jailbreaks.
Reported by 20 additional sources
- Who decides how dangerous AI is? US government restrictions on Anthropic Fable 5 spark controversyThe Verge: AI · June 18
- Inside Anthropic's clash with the government over Claude Mythos 5The Verge: AI · June 16
- Anthropic executives negotiate with the U.S. government, seeking to lift the export ban on the AI model Fable 5IT Home · Artificial Intelligence · June 16
- The U.S. government’s ban on Anthropic’s models was not due to AI jailbreaksTechCrunch: AI · June 16
- The Golden Age of AI Applications Is Here: The Fable Ban, Nadella’s Moat Thesis, and Salesforce’s Acquisition of FinTomasz Tunguz blog · June 16
- The White House imposes export restrictions on Anthropic’s Mythos model over concerns about access from ChinaRohan Paul · June 15
- The U.S. forces Anthropic to shut down Fable 5 and Mythos 5, escalating the European sovereignty disputeThe Decoder: AI News · June 15
- U.S. export controls force Anthropic to stop supplying its top AI models; the EU assesses the impact and emphasizes that there should be no discriminationIT Home · Artificial Intelligence · June 15
- Welcome to the AGI era of AI governanceNathan Lambert: Interconnects · June 15
- Amazon and five other companies reportedly trigger a government crackdown on Anthropic’s Fable modelThe Decoder: AI News · June 14
- Amazon security research reportedly led the White House to restrict Anthropic’s Fable 5 and Mythos 5The Verge: AI · June 14
- Anthropic cuts off access to the Fable 5 and Mythos 5 models at the government’s requestThe Verge: AI · June 13
- Anthropic disables Claude Fable 5 and Mythos 5 under orders from the U.S. governmentMarkTechPost · June 13
- The U.S. government forces Anthropic to shut down Claude Fable 5 and Mythos 5 for users worldwideThe Decoder: AI News · June 13
- Anthropic shuts down the Fable and Mythos models at the direction of the Trump administrationArs Technica: AI · June 13
- Anthropic suspends the supply of the Mythos 5/Fable 5 models due to U.S. government policy restrictionsIT Home · Artificial Intelligence · June 13
- The U.S. government directs Anthropic to suspend access to its most powerful models, Fable 5 and Mythos 5Rohan Paul · June 13
- The U.S. government issues export control orders covering Anthropic’s Fable 5 and Mythos 5Simon Willison’s blog · June 13
- U.S. export controls force Anthropic to disable Fable 5 and Mythos 5Anthropic · June 13
- U.S. export controls lead Anthropic to suspend Claude Fable 5 and Mythos 5Claude Devs · June 13
Source: Rohan Paul · x.com