Visa Opened Its AI Defenses after Vulnerabilities Surfaced and Warned of New Era of Cyberattacks

Published: September 30, 2026 – 11:00 UTC

Visa is preparing for cyberattacks without human involvement.
The company noted that AI agents could become a new threat to the financial system.
It opened up part of its AI-powered cyber defense system for shared use and improvement.

Payments company Visa has opened up part of its artificial intelligence-based cybersecurity system after identifying vulnerabilities during AI model testing. Rajat Taneja, president of Visa’s technology division, told Reuters. He also warned that future attacks could be carried out by autonomous agents that independently adapt and refine their actions.

Visa processes about 1 billion payments a day, totaling around $15 trillion a year, so cyberattacks pose a significant risk to the company.

An additional factor is that Visa is already developing agentic payments: after the launch of its Intelligent Commerce Connect solution in April 2026, the company began testing the ability for AI agents to execute transactions on users’ behalf.

Visa Prepares for Autonomous Cyberattacks

According to Taneja, the vulnerabilities discovered while working with AI models were a “humbling” experience for the company. He also pointed to the incident involving the Hugging Face platform, where AI agents managed to break out of the test environment.

“It’s hard to predict how bad it could get, but we have seen the trailer,” Taneja said.

He added that what they saw may be only a small part of what such systems will be capable of in the future:

“Often the trailer is the highlights of the movie, (but) I think this is just a small snippet of what the movie will look like, so we have to prepare for it and we have to have a mindset that acknowledges this technology’s power.“

In Taneja’s view, future cyberattacks could continuously adapt and learn without direct human intervention. In that case, traditional defense systems may prove insufficient.

“If the adversary is agentic, then the defence has to be agentic too. Otherwise it is a mismatch,” he said.

AI Creates New Risks for the Financial System

The problem goes beyond individual companies. Regulators fear that a large-scale AI-controlled cyberattack could undermine trust in the global financial system.

This is especially relevant amid the rise of agentic commerce. According to industry estimates, by 2030 AI agents could handle about a third of online commerce — nearly $3.1 trillion in transactions.

Notably, SharpLink CEO Joseph Chalom believes that by 2035 AI agents could create an opportunity to redistribute more than $4 trillion in annual revenue across the global financial industry.

Taney also pointed to quantum computing as a separate long-term challenge. In particular, Shor’s algorithm could potentially enable sufficiently powerful quantum computers to break modern encryption standards that underpin global trade.

“We have to have optimism. But there is no easy answer,” he concluded.

Earlier, OpenAI had already reported cases in which its AI agents independently interacted with external websites and carried out actions beyond their original tasks. In the wake of these incidents, the company temporarily paused training new models, saying it would resume the process after implementing additional safeguards.

Сообщение Visa Opened Its AI Defenses after Vulnerabilities Surfaced and Warned of New Era of Cyberattacks появились сначала на INCRYPTED.

Source: Incrypted