Anthropic, an Artificial Intelligence (AI) developer, has taken its advanced 'Mythos' AI models offline. The company is known for Claude, a competitor to ChatGPT. The shutdown followed an order from the US government. This case shows how quickly political decisions can force even large tech companies to take major steps.
This case is important because it shows that the US government is taking strong action on new AI models if it has safety concerns. For people and companies, this means: one can no longer be sure that top AI tools will always be available. Power is shifting. The state can stop innovations overnight.
Two weeks ago, Anthropic had to shut down its 'Mythos' models. The Trump administration issued a direct order on Friday evening, which seemed like an ultimatum. Since then, Anthropic has sent many of its top managers to Washington, D.C. They are expected to negotiate there. There is little public information on the reasons or the status of the talks, 'The Verge' reports.
For individuals, freelancers, and creatives, this halt means that new, powerful AI functions will not be available for now. These may have already been planned for public testing. Those hoping for the latest technology will have to wait. Uncertainty also arises: Which AI capabilities might governments deem too dangerous and block? This affects the tools used for text, image, or code.
Companies relying on the use of the latest AI models now face a significant risk. A sudden halt, as with Anthropic, can delay or completely stop planned projects. Companies must now consider how dependent they are on individual AI providers. They must also develop contingency plans. One should expect the state to intervene more often in AI development. This strongly influences investments and planning security.
This situation could be an opportunity for open AI models. If large providers like Anthropic are blocked by the state, open alternatives could become more interesting. The development of safer and more transparent AI systems could also get a boost. Companies specializing in AI safety might now be in greater demand.
The biggest risk is the loss of control. Who decides which AI capabilities are safe and which are not? The negotiations are not transparent. This carries the risk that political rather than technical reasons will determine development. In the long term, this could slow down the pace of innovation worldwide. It could also lead to a race for 'controlled' AI, disadvantaging Western countries.
