What We Know
Multiple outlets report that Anthropic and the U.S. government have moved toward a deal that will lift some curbs on the company’s latest AI model, Mythos 5. Commerce Secretary Howard Lutnick has exempted “certain trusted partners” from export-license requirements for Mythos 5, a carve‑out that reportedly covers named entities and their foreign‑national employees. After roughly two weeks of globally suspended access, Mythos 5 was cleared for use by more than 100 U.S. institutions, according to reporting.
Coverage describes the change as a partial loosening rather than a full reversal: Mythos 5 has been the primary beneficiary so far, while Anthropic’s Fable 5 remains subject to export controls. Some reports indicate limited, managed releases of Mythos and Fable models to select parties, but the precise terms vary across accounts. The moves have been described in multiple outlets as a decision by the current administration to partially lift prior restrictions on Anthropic’s most advanced AI offerings.
Source Comparison
Aligned reportingCorroborates
- Bloomberg↗Reports that Anthropic and the U.S. moved toward a deal to lift curbs on its AI models and describes a partial loosening that chiefly benefits Mythos 5.
- news.bloomberglaw.com↗Reports a U.S.–Anthropic move to ease export controls, describing carve-outs for certain trusted partners and limited reinstatements of access to Mythos 5.
- aiweekly.co↗Notes Commerce Secretary Howard Lutnick exempted “certain trusted partners” from export-license requirements for Mythos 5, including named entities and their foreign‑national employees.
- dnyuz.com↗States the U.S. loosened restrictions on Anthropic’s Mythos AI model, aligning with accounts of a partial lifting of prior export curbs.
- iowapublicradio.org↗Reports the current administration partially lifted the export ban on Anthropic’s most advanced AI model, consistent with descriptions of a partial easing rather than a full reversal.