Connect with us

Hi, what are you looking for?

AI Business

Anthropic’s Mythos Raises Cybersecurity Concerns as Major Banks Seek Access

Anthropic’s Mythos, now in select use by JPMorgan and other major banks, sparks urgent cybersecurity debates as European regulators assess its unprecedented threat.

The emergence of Anthropic’s Mythos is prompting a scramble among major banks to access and test the technology, while regulators rapidly assess the cybersecurity risks it poses. At last week’s International Monetary Fund spring meeting in Washington, concerns were raised about how prepared financial institutions are to manage the challenges presented by this new AI model. Several US banks have already begun utilizing Mythos, placing pressure on the rest of the industry to catch up.

“It’s certainly not something that’s causing panic or setting off any alarm bells on our end right now, but it’s definitely something we need to keep in mind in our day-to-day risk management — and that’s exactly what we’re doing,” said Christian Sewing, CEO of Deutsche Bank. He noted that banks are in close communication with European regulators regarding Mythos. Sewing emphasized that banks have their own contingency plans in place, and while everyone is eager to gain access, he believes it is appropriate for access to be limited for now.

Anthropic has restricted access to Mythos primarily to partners in its Project Glasswing initiative, along with around 40 additional organizations that are involved in managing critical software infrastructure. JPMorgan is the only bank publicly acknowledged by Anthropic to have access, though Bank of America has been part of the initiative from the beginning and is testing Mythos technology internally, according to a source familiar with the matter.

Recently, other US banks have reported gaining access to Mythos. Morgan Stanley CEO Ted Pick confirmed during the bank’s earnings call that the bank is involved with the Claude Mythos Preview, stating that discussions about cyber risks are ongoing within the Financial Services Forum. He acknowledged the increasing threat posed by cyber risk and expressed optimism that collaboration will lead to better security measures. Goldman Sachs CEO David Solomon also noted during his earnings call that the bank has access to Mythos, asserting the institution’s commitment to working closely with Anthropic to leverage its capabilities.

Additionally, Citigroup has been reported to have access to Mythos for internal testing, according to an informed source. Some banks that have not gained access are questioning the equity of the current access model, particularly whether JPMorgan has an unfair advantage. This discussion is expected to be brought up with the US Treasury; both the Treasury and Anthropic did not respond to requests for comment.

Multiple senior banking and regulatory sources in Europe have indicated that they are unaware of any European financial institutions currently having access to Mythos. The British government sent an open letter to Anthropic leaders on April 15, citing that testing by its AI Security Institute demonstrated Mythos’s capabilities in cyber offense to be “substantially more capable” than any models previously assessed.

During the IMF meetings, Mythos was a recurring topic. European supervisors are currently evaluating the model through existing cyber resilience frameworks and have not yet expressed alarm. However, the high-level coding capabilities of Mythos are believed to afford it an unprecedented ability to identify cybersecurity vulnerabilities, drawing heightened scrutiny from global regulators.

CS Venkatakrishnan, CEO of Barclays, stated that Mythos represents a serious threat to the global banking system and predicted that it would likely give rise to similar, even more powerful cyber threats. As the banking industry continues to respond to the challenges posed by Mythos, its implications for cybersecurity and risk management are expected to reverberate across the sector. — Tom Sims, Saeed Azhar, Tatiana Bautzer, and Jesus Aguado, with Francesco Canepa, Balazs Koranyi, and Lawrence White in London, (c) 2026 Reuters

See also
Marcus Chen
Written By

At AIPressa, my work focuses on analyzing how artificial intelligence is redefining business strategies and traditional business models. I've covered everything from AI adoption in Fortune 500 companies to disruptive startups that are changing the rules of the game. My approach: understanding the real impact of AI on profitability, operational efficiency, and competitive advantage, beyond corporate hype. When I'm not writing about digital transformation, I'm probably analyzing financial reports or studying AI implementation cases that truly moved the needle in business.

You May Also Like

AI Technology

Victory Giant Technology Huizhou's shares soared 59.6% on their Hong Kong debut, raising $2.2 billion to expand production amid China's semiconductor push.

AI Cybersecurity

NSA accesses Anthropic's Mythos AI for cybersecurity vulnerabilities, despite Pentagon's blacklist, highlighting urgent national defense implications.

AI Regulation

Indonesia aims for an 8.35% growth in its ICT sector by 2025, positioning itself as a key player in global AI governance while addressing...

Top Stories

Google DeepMind forms a dedicated team led by Sebastian Borgeaud to enhance AI coding models, aiming to close the 50% coding gap with Anthropic's...

AI Cybersecurity

Anthropic's Claude Mythos can detect software vulnerabilities in minutes, exposing South Africa’s cybersecurity readiness gap as 77% of firms take over a week to...

AI Regulation

NVIDIA announces a $40 billion diversified AI foundation portfolio, strategically addressing missed investments in OpenAI and Anthropic while boosting shares by 98.8%.

Top Stories

OpenAI acquires personal finance startup Hiro and media company TBPN to bolster talent and improve public image amid fierce competition from Anthropic.

AI Regulation

Anthropic's Claude Mythos launches with minimal EU regulatory input, raising alarms as concerns grow over unregulated AI amid a $300M pro-AI campaign in the...

© 2025 AIPressa · Part of Buzzora Media · All rights reserved. This website provides general news and educational content for informational purposes only. While we strive for accuracy, we do not guarantee the completeness or reliability of the information presented. The content should not be considered professional advice of any kind. Readers are encouraged to verify facts and consult appropriate experts when needed. We are not responsible for any loss or inconvenience resulting from the use of information on this site. Some images used on this website are generated with artificial intelligence and are illustrative in nature. They may not accurately represent the products, people, or events described in the articles.