• Anthropic is considering designing its own AI chips to address processor shortages, though plans remain in early stages and no final decision has been made.
  • The company’s revenue has surged to over $30 billion while its new Mythos AI model prompted urgent cybersecurity warnings from US officials to major banks.

The artificial intelligence sector is entering a new phase as companies are racing to secure the computing power needed for their most advanced systems.

Anthropic, the maker of the Claude AI assistant, is now considering joining tech giants in designing its own specialized chips, while its latest AI model raises serious cybersecurity concerns across government and industry.

Chip Design Plans Take Shape

Anthropic is trying to design custom AI chips, as the move comes after the company and its competitors are facing ongoing shortages of processors that are needed to build and run advanced AI systems. ⁽¹⁾

However, the plans remain in early stages, as the company has not yet confirmed its specific chip design for the project.

This exploration mirrors similar efforts at other major AI companies, including Meta and OpenAI. Designing an advanced AI chip typically costs around half a billion dollars, according to industry sources. ⁽²⁾

Revenue Growth Drives Computing Needs

The potential chip project comes as demand for Claude has surged dramatically. Anthropic announced this week that its revenue run rate now exceeds $30 billion, up sharply from approximately $9 billion at the end of 2025. ⁽³⁾

Run rate is a financial metric that projects a company’s future performance by using current financial data

Currently, Anthropic relies on a mix of chips from different suppliers. The company uses tensor processing units designed by Google and chips from Amazon to develop and operate Claude.

Earlier this week, Anthropic signed a long-term agreement with Google and Broadcom, which helps design the TPUs. The deal supports Anthropic’s commitment to invest $50 billion in strengthening US computing infrastructure. ⁽⁴⁾

Cybersecurity Warnings Over New Model

Beyond chip design, Anthropic faces a different challenge with its latest AI model. US Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell called an urgent meeting with bank CEOs this week to discuss cyber risks posed by the new Mythos model.

Anthropic launched Mythos earlier this week but limited its release due to concerns about cybersecurity vulnerabilities. The company says the model can identify and exploit weaknesses across every major operating system and web browser. ⁽⁵⁾

The Treasury-hosted meeting in Washington took place on Tuesday. Its purpose was to ensure banks understand the risks from Mythos and similar models and are taking steps to protect their systems. ⁽⁶⁾

Anthropic has been in ongoing discussions with US government officials about the model’s offensive and defensive cyber capabilities.

Software Sector Feels the Impact

The software industry is feeling pressure from Anthropic’s latest developments. Software stocks have declined again this week.

The decline follows Anthropic’s disclosure of its $30 billion revenue run rate and its rollout of new agentic tools. These include Claude Managed Agents, which reduce the time developers need to build their own AI agents. The announcements have revived fears that AI could disrupt traditional software companies. ⁽⁷⁾

Sources: ⁽¹⁾ ⁽²⁾ CNBC, ⁽³⁾ ⁽⁴⁾ Wall Street Journal, ⁽⁵⁾ ⁽⁶⁾ ⁽⁷⁾ Reuters