Anthropic's Claude Fable 5 Banned by Trump Administration Amid Comparison with OpenAI's GPT 5.5
Anthropic recently released its Claude Fable 5 AI model, a powerful public version of its unreleased Mythos model. Days after its launch, the Trump administration issued an export control directive, banning Fable 5 for foreign nationals. Anthropic subsequently disabled the model for all users to comply with the order. Anthropic stated its belief that the order aimed to facilitate 'jailbreaking' Fable 5 to identify vulnerabilities, claiming other models like OpenAI's GPT 5.5 could achieve similar results. Despite comparable feature sets as large language models, Fable 5 demonstrated superior performance on several AI leaderboards and benchmarks compared to GPT 5.5.
Earlier this week, Anthropic launched Claude Fable 5, its most powerful public artificial intelligence model. Fable 5 is described as a "safe" version of the company's more powerful, unreleased Mythos model, which remains private due to potential dangers.
Shortly after its release, the Trump administration issued an export control directive that prohibited the use of Fable 5 by foreign nationals. This order compelled Anthropic to disable Fable 5 for all its customers to ensure compliance.
Anthropic issued a statement suggesting that the U.S. government's order was intended to facilitate a method to "jailbreak" Fable 5 and identify its vulnerabilities. The company further claimed that other AI models, including OpenAI's GPT 5.5, could achieve the same results without needing a bypass.
Both Claude Fable 5 and GPT 5.5 are large language models designed for tasks such as document analysis, data interpretation, and advanced coding and development, rather than image or video generation. They are considered direct competitors in their use cases.
On various AI leaderboards, Claude Fable 5 demonstrated significant performance leads. It secured the top position on the popular Arena leaderboard, with Anthropic's Claude Opus 4.7 Thinking and Claude Opus 4.8 Thinking ranking second and third, respectively. OpenAI's GPT 5.5 was placed fourth. Fable 5 also led rankings on Artificial Analysis, Simple Bench, and other benchmark leaderboards.
According to Anthropic's own comparison data, Fable 5 outperformed GPT 5.5 across multiple benchmarks, including agentic coding, knowledge work, and cybersecurity, at the time of its launch.
Prior to the export control order, Fable 5 was available to Anthropic's paid subscribers on Pro, Max, Team, and seat-based Enterprise plans at no additional cost until June 22. Anthropic had planned to transition Fable 5 to a usage-based paid add-on model for these plans starting June 23, with pricing on its Claude API and consumption-based Enterprise plans set at $10 per million input tokens and $50 per million output tokens. Currently, Fable 5 is unavailable to all Anthropic users, while OpenAI's GPT 5.5 remains accessible to OpenAI Plus, Pro, Business, and Enterprise users via ChatGPT.
(Source: Mashable Tech)
