The Australian Securities and Investments Commission ( ASIC ) has said it is closely monitoring the development of Anthropic’s new artificial intelligence model Mythos , expressing concerns about its potential impact on the financial system . The statement comes as regulators around the world are scrambling to address what is considered one of the most powerful and controversial AI systems to date.
ASIC said it was working with other regulators and the financial industry to understand and manage new technological risks, and stressed that financial service providers must be at the “front line” of protecting their customersin an environment that is rapidly changing due to artificial intelligence.
Mythos and Project Glasswing: An AI with “offensive security” capabilities
Mythos was unveiled in early April 2026 as part of Project Glasswing, a limited-access program that has generated both interest and concern. According to Anthropic, the model has the ability to detect and exploit zero-day vulnerabilities in operating systems and browsers, with the aim of improving cybersecurity.
See also: Anthropic AI Mythos Preview: Helps Apple Find Vulnerabilities
However, this capability has raised concerns among regulators, who fear that in the wrong hands it could become a tool for large-scale cyberattacks. The debate is now not just about technological progress, but also about whether such systems can remain under safe control.
Global alarm to central banks and regulators
The concern is not limited to Australia. The governor of the Bank of England warned that Mythos could “blow up the cyber risk ecosystem,” calling for an immediate assessment of the impact on financial infrastructure. Similarly, the European Central Bank admitted that there is not yet an adequate regulatory framework to deal with such technologies.
In the United States, the Treasury Department and the Federal Reserve have called an emergency meeting with top Wall Street to discuss potential cyber threats. The fact that major financial institutions are actively participating in the discussion shows the extent of concern about the new model.
The involvement of banks and the first tests
Despite the fears, major US banks have already begun testing Mythos in controlled environments , with Goldman Sachs and JPMorgan Chase among the first financial institutions to gain access to the system, recognizing both its potential and its risks .
Project Glasswing includes around 40 technology and financial partners, including giants like Amazon, Microsoft and Google, creating a particularly complex development and testing ecosystem.
See also: Kuwaiti Banks Strengthen Threat Response
Structural risks to the global financial system
Regulators point out that the main risk comes not from isolated incidents, but from the structure of the financial system itself. The coexistence of legacy infrastructures with modern cloud environments creates a complex technological landscape, full of unseen security gaps.
The concentration of critical services in a few large cloud providers adds to the concern. A powerful AI system that identified vulnerabilities in these providers could theoretically affect much of the global financial system at once, creating a chain reaction.
Controversy over access and transparency
Anthropic has adopted policy restricted access for Mythos, arguing that the controlled availability of the technology reduces the risk of abuse. However, this approach has been criticized by industry experts, who argue that transparency and broader control are more effective security tools.
The company is also facing legal and regulatory challenges, as it is embroiled in a dispute with the US Department of Defense over national security issues and collaborations with government agencies.
🔒 Protect your privacy with Proton VPN
Swiss VPN from the creators of Proton Mail — strict no-logs policy, strong encryption, and built-in NetShield that blocks ads, trackers, & malware.
- ✔ No-logs, based in Switzerland (except 14-Eyes)
- ✔ NetShield: blocks ads, trackers & malicious domains
- ✔ Covers all devices — free version available
The link is an affiliate link — SecNews may receive a commission at no additional cost to you. It does not affect the independence of our article writing.
See also: 6 new Android malware targets banking apps
The stakes of the next day in AI and finance
Mythos is not just another technological step, but a potential catalyst for profound changes in the way we understand cybersecurity. The ability of a model to identify and exploit vulnerabilities in real time redefines the boundaries between defense and offense.
As governments, banks and tech companies try to adapt, it is becoming clear that the next phase of artificial intelligence will not be judged just by the power of the models, but by how effectively they can be controlled.
