On the evening of March 2, millions of users relying on Claude were left in shock as the software encountered errors and became unusable. The outage coincided with an explosion at a data center in the UAE and subsequent actions by the U.S. Treasury.

Claude, developed by Anthropic, is a strong competitor to ChatGPT and backed by Nvidia. The unexpected events raise questions about whether they are mere coincidences or part of a deliberate attack.

Government Shutdown of AI Company

One of the most shocking tech news recently is the U.S. government’s decision to shut down its own AI star company, Anthropic. Despite being a homegrown American company that hasn’t committed any legal violations, the government retaliated against Anthropic for refusing to comply with Pentagon demands to turn its AI technology into military tools.

A few days prior, the Pentagon approached Anthropic, asking them to open unlimited access to Claude for military applications, including intelligence analysis and target recognition. Anthropic, which emphasizes ethical AI and safety, rejected the request outright.

This refusal angered the Trump camp, leading to the official announcement of the shutdown. U.S. Treasury Secretary Scott Bessent declared that all federal agencies must cease using Claude’s products, giving them only a six-month grace period to transition.

Agencies like the Federal Housing Finance Agency and Fannie Mae swiftly removed Claude from their systems overnight. In an unprecedented move, the government labeled Anthropic as a “supply chain risk,” a term usually reserved for foreign companies, marking the first time it was applied to a domestic firm.

AI’s Vulnerability Exposed

At the same time as Anthropic faced government shutdown, Claude experienced a global outage, affecting millions of users. On March 2 at 7:49 PM, many employees at major companies attempted to use Claude for last-minute tasks, only to encounter error messages and unresponsive screens. Initially, it was thought to be a server issue, but by 10:35 PM, the core model, Claude Opus 4.6, faced significant errors, followed by another model, Claude Haiku 4.5, at 11:56 PM.

The outage lasted nearly 10 hours, finally returning to normal at 5:16 AM the next day. The cause was linked to an explosion at an Amazon AWS data center in the UAE, which was hit by an unidentified object shortly after the errors began.

The incident highlighted the fragility of the global AI infrastructure, revealing how reliant Claude is on AWS’s computing power. This vulnerability underscores the geopolitical risks that can disrupt digital operations.

Silicon Valley’s Response

Silicon Valley companies previously believed that strong technology and capital would allow them to thrive without government interference. However, Anthropic’s situation has changed that perception. As a prominent company backed by Nvidia’s substantial investment, Anthropic’s fate has instilled fear in others.

In an unprecedented show of solidarity, competitors like OpenAI, Google, and IBM joined forces, with over 600 employees and industry leaders signing a letter urging the government to retract the “supply chain risk” designation against Anthropic and calling for congressional review.

They expressed concerns that if companies face retaliation for not complying with contracts, innovation would suffer. The fear is not just for Anthropic but for the future of all companies in the face of government power.

Interestingly, while Anthropic faced backlash, OpenAI seized the opportunity to announce a partnership with the U.S. Department of Defense, integrating ChatGPT into military operations. This prompted a backlash from users, leading to a “QuitGPT” movement, with many uninstalling ChatGPT in protest.

Self-Inflicted Wounds

The U.S. government’s labeling of Anthropic as a “supply chain risk” suggests potential supply disruptions, particularly affecting Nvidia, which provides essential computing power to Anthropic. Nvidia’s CEO now faces a dilemma: if they comply with the government and cut off supplies, they risk losing a lucrative deal with Anthropic, which recently raised over $600 billion, half of which was just received for chip procurement.

This situation exemplifies the self-destructive nature of U.S. policy, aiming to control the tech landscape while inadvertently harming its own companies. The perception of Silicon Valley as a haven for innovation is now threatened, as government actions can disrupt domestic enterprises.

As a result, regions like Europe, the Middle East, and China are accelerating efforts to establish their own computing infrastructures. The U.S. government’s politicization of technology may drive global efforts to reduce dependence on American systems.

Ultimately, Claude’s predicament signifies more than just a single company’s misfortune; it marks the beginning of a significant shift in the global AI landscape. When technology is forced to bow to power, the so-called “golden age of AI” may be overshadowed by emerging challenges.