Safety

Three major Japanese banks likely to gain access to Anthropic's Mythos AI model within weeks

Three of Japan's largest banks are expected to obtain access to Anthropic’s Mythos AI model within about two weeks, raising cybersecurity concerns for the domestic financial sector.

Three of Japan’s largest banks — Micubisi UFJ Financial Group, Mizuho Financial Group and Szumitomo Micui Financial Group — are expected to gain access to Anthropic’s Mythos artificial intelligence model within about two weeks, according to a Reuters source familiar with the matter.

The arrival and wider availability of Mythos have prompted significant cybersecurity concerns within the financial sector. Cybersecurity experts warn the model carries substantial risks, especially for institutions that operate outdated IT systems. Regulatory authorities have already issued multiple warnings related to these risks.

Several banks in the United States have already been granted access to the Mythos system, and Anthropic has indicated plans to expand the service to European and British banks. Japan appears to be the next market in line.

None of the affected Japanese banks commented on the report, and Anthropic did not respond to requests for comment.

Government response and working group

Katajama Szacuki, Japan’s finance minister, met on Tuesday with Scott Bessent, identified in the report as the U.S. finance minister. Following the meeting, Szacuki announced that Japan will establish a new working group this week. The panel will include both public and private sector participants and will be tasked with addressing the cybersecurity risks Mythos may pose to Japan’s financial system.

The working group’s first meeting is scheduled for Thursday.

Why this matters

  • Providing banks with access to Mythos could accelerate adoption of the technology in financial services while also introducing new attack surfaces and vulnerabilities.
  • Institutions running older IT infrastructure are viewed as particularly exposed, since their defenses are often weaker.
  • Regulatory warnings and the formation of a government-led working group indicate that policymakers are taking the security implications of the technology seriously.

Next steps

According to Reuters, the working group will hold its initial meeting on Thursday. The specific measures and any regulatory guidance will depend on the panel’s findings and recommendations.

Source: Reuters

Lead image for illustration only.