By MyEngineers
The United States government has reversed an export ban that had blocked access to Anthropic’s most powerful artificial intelligence tools, just weeks after imposing the restriction over national security worries.
Anthropic said in a statement that it will start restoring access to two of its AI models, Claude Fable 5 and Mythos 5, from Wednesday, after the US Department of Commerce notified the company that the restrictions had been lifted.
These two models are Anthropic’s most advanced AI tools. They were suddenly suspended on 12 June following concerns that the technology could be exploited by hackers to find and take advantage of weaknesses in computer systems.
In a letter seen by the BBC, the Commerce Department said Anthropic had taken steps to address the risks that led to the ban. Commerce Secretary Howard Lutnick wrote that Anthropic agreed to proactively detect and address security risks tied to the models.
According to Lutnick, the company also agreed to work with the government on future releases of its AI models and to report any malicious activity linked to the technology. He added that the Commerce Department still reserves the right to bring back the restrictions if it becomes necessary.
The BBC said it has reached out to the Department of Commerce for further comment.
Mythos and Fable are built on Anthropic’s Claude platform, which competes with OpenAI’s ChatGPT and Google’s Gemini. Fable 5 is designed for everyday consumers and can carry out deep reasoning and complex tasks on its own. Mythos 5, on the other hand, is built for businesses and cybersecurity professionals, and it is reportedly able to identify weaknesses in computer code and exploit them. Both models were launched on 9 June.
Anthropic had earlier said that US authorities did not point to any specific problem with its technology even while ordering the suspension of both platforms worldwide. The company said at the time that the government appeared to have found a way of bypassing or jailbreaking Fable 5, but it disagreed that a narrow jailbreak discovery should be reason enough to pull back a commercial product already in use by hundreds of millions of people.
Why It Matters
This case is worth watching for Nigeria and other developing countries that are just beginning to explore how far artificial intelligence can go, both in business and in national security matters.
It shows clearly that even the biggest AI companies in the world are not fully in control of how their products are regulated. For Nigeria, this is a lesson in how important it is to build local regulatory capacity around AI, rather than depending entirely on foreign companies or foreign governments to decide what tools Nigerian businesses, banks, and government agencies can or cannot use.
It also raises a bigger question about dependency. Nigerian companies, from fintech firms to telecoms to government ministries exploring AI adoption, are increasingly relying on foreign built AI systems like Claude, ChatGPT, and Gemini. If a foreign government can suspend access to these tools over its own national security concerns, Nigerian users could suddenly find themselves locked out of services they depend on, with no say in the matter at all.
This is part of the argument for why Nigeria needs to invest more seriously in building its own AI capacity, whether through local research institutions, universities, or partnerships that allow for technology transfer rather than pure dependency.
The cybersecurity angle also matters. If a tool like Mythos 5 can identify and exploit weaknesses in computer code, that is a double edged sword.