Claude crashed: massive failure amid geopolitical pressure on Anthropic
On June 18, the Claude chatbot from Anthropic experienced a major outage. Thousands of users worldwide encountered the service ceasing to respond and failing to load messages. According to monitoring systems, the peak of complaints occurred at 10:03 Moscow time.
This incident occurred during an extremely tense period for the company. This week, Anthropic found itself at the center of several events: from a major bank blocking access to its models to negotiations with US and EU authorities.
Scale of Technical Issues
Over the course of the day, the number of complaints about Claude's operation ranged from 420 to over 770. Users reported that messages were not being sent or loaded in the browser, and the service itself simply was not responding. The monitoring service StatusGator recorded signs of the outage around 10:00 Moscow time, noting that the problem remained without official confirmation for some time. Some free-tier users received a notification that Claude could not respond due to insufficient capacity. This is not the first service disruption this month: in early June, Claude already experienced a major outage affecting the web interface, mobile app, and the Claude Code platform.
Geopolitical Background: JPMorgan and G7
The outage coincided with serious pressure on the company. JPMorgan Chase banned its employees in Hong Kong from accessing Anthropic's models, repeating a move by Goldman Sachs, which excluded Claude from the list of approved tools for its bankers back in April. These decisions fit into the overall rise in tensions between the US and China over AI technologies, data security, and access to advanced computing tools.
The situation is further aggravated by the US export directive. US Commerce Secretary Howard Lutnick ordered Anthropic to suspend the export of the Mythos and Fable models to all foreign nationals, citing the risk of their use by the military intelligence of China, Russia, and other countries. Against this backdrop, a G7 leaders' dinner took place on June 17 with the participation of Anthropic CEO Dario Amodei, and a meeting between the EU cybersecurity agency ENISA and Anthropic was scheduled for June 18.
Restrained EU Reaction
Despite Washington's tough stance, Europe is not inclined towards confrontation. Diplomats and officials from G7 countries emphasize that they are ready to jointly reduce AI risks with the US and want to turn the incident into a platform for cooperation, rather than a reason for division. The EU had hoped to achieve unity on advanced models before the summit's end, with the main task being to restore trust. Notably, the report on the bilateral meeting between European Commission President Ursula von der Leyen and Trump on the sidelines of the summit did not mention the Anthropic issue at all. Brussels is acting cautiously also because no official notification of export controls has been received from US authorities—only a statement from Anthropic itself exists.
Cryptalist Analysis: Technical failures in Claude's operation are just the tip of the iceberg. Anthropic has found itself in the crossfire of geopolitics and corporate interests. The blocking by JPMorgan and Goldman Sachs is not just a technical decision but a market signal that trust in the company's models has been undermined. In the short term, this could lead to a loss of market share in the corporate segment, especially in Asia. The AI market is becoming a battlefield not only of technologies but also of politics, and Anthropic is one of the key targets here.