Too powerful to release? Why Anthropic’s Mythos is raising global alarm

A new artificial intelligence model developed by Anthropic is triggering growing concern among governments, businesses and analysts, with its advanced capabilities in cybersecurity and software engineering seen as both groundbreaking and potentially disruptive.

The model, dubbed Mythos, has already been restricted from wider release due to its powerful features, which the company itself admits are pushing beyond human limits in some areas.

“We have a model that's beginning to outstrip human capabilities in the cyber world,” said Guillaume Princen, Anthropic’s Paris-based chief of relations with startups and tech firms.

Highlighting its unprecedented ability, he added, “Mythos is capable of spotting security holes that have existed for decades, in systems tested by both human experts and automated tools, that have never been discovered before.”

Restricted rollout amid rising concerns

Anthropic has so far limited Mythos’ access to a select group of major firms such as Nvidia, Amazon, Apple and JPMorgan Chase, allowing them to test and strengthen their systems.

The company says this cautious approach is deliberate.

“We prefer to be transparent and lay these risks out on the table,” Princen said, adding that AI safety concerns are “central to Anthropic's DNA”.

“We don't have all the answers, this has to be a conversation between tech actors like us who have the data, the academic world, the political world and the world of economists,” he added.

However, the model’s capabilities have unsettled regulators and policymakers. The British government warned that Mythos “highlights the speed at which AI capabilities are increasing and the threats they potentially pose”, while the European Union has sought further clarity.

“Mythos is certainly not a model that will soon be opened to the public at large, for obvious reasons,” Princen said, underlining concerns about its misuse.

Cybersecurity breakthrough or a new threat?

While Mythos can strengthen defences by identifying vulnerabilities, experts warn the same capabilities could be exploited for offensive cyber operations if misused. The fact that no European firms are part of Anthropic’s “Project Glasswing” initiative has also raised questions about global preparedness against such advanced AI tools.

At the same time, Anthropic faces criticism that it may be overstating the model’s capabilities amid intense competition with OpenAI.

Impact on India’s IT sector

Beyond cybersecurity, Mythos could reshape the global technology services industry. A report by Kotak Institutional Equities warned that such advanced AI systems may pose near-term disruption risks to India’s IT sector.

The report said the model “exhibits a step-jump in benchmark performance across software engineering tasks” and added that it “raises near- to medium-term disruption risks for IT services,” particularly for companies heavily dependent on application development.

Highlighting the scale of potential impact, it noted, “The realisation of similar improvements in real-world scenarios risks turning our estimate of a 3-3.5% annual growth headwind for the industry... from prudent to practical.”

Kotak further warned that such advancements could “increase efficiencies across all IT services segments” but may also deepen inequalities across the sector and trigger pricing pressure.

It said the development “could pressurise the valuation multiples of IT services companies” and compound “near-term deflation risks for services.”

However, the report also pointed to opportunities, noting that AI could “accelerate GenAI-driven business use cases, providing large new opportunities to Indian IT.”