Hackers use AI to create first zero day exploit that scam 2FA on a global scale

The Google Cybersecurity Action Team announced the identification of a criminal group that used a zero-day exploit to bypass large-scale two-factor (2FA) authentication. For the first time, it is believed that vulnerability and exploit were generated with the aid of artificial intelligence, marking a dangerous leap in hackers' ability to automate attacks against critical security systems.

How did the AI create the zero-day exploit that scams 2FA?

According to Google's report, hackers used AI generative models to analyze the 2FA system code widely adopted by companies. The AI identified an unknown vulnerability (day zero) and then automatically generated the exploit code capable of bypassing verification in two stages without direct human interaction. This process, which would require months of manual expert work before, was completed in days.

2FA systems affected by the AI exploit

The attack targeted corporate multifactor authentication solutions based on TOTP (Time-based One-Time Password) tokens and push notifications. Exploit exploits a server-side validation failure, allowing the intruder to intercept and replicate verification codes without user knowledge. Google did not disclose names of specific suppliers, but stated that vulnerability has already been corrected in emergency updates.

Why is the AI explosion a global security alert?

This case is the first documented example of AI being used not only to automate existing attacks but to discover new vulnerabilities and create functional exploits. Experts fear that the technique will spread, drastically reducing the cost and time to develop cyber weapons. The incident reinforces the urgency of AI regulations, as discussed in the Anthropic Mythos cybersecurity crisis.

What should Brazilians do against AI explosion in 2FA?

Security teams need to adopt a proactive stance: immediately update all 2FA systems, monitor authentication logs in search of abnormal patterns and review incident response policies. In addition, it is recommended to evaluate phishing-resistant authentication solutions such as FIDO2 security keys, which offer greater protection against automated exploits.

Frequently Asked Questions about zero day exploit with AI

What's a zero-day exploit?

It is a security vulnerability unknown by the developer and uncorrected, which can be exploited by attackers before a patch is released.

Can the AI be used for defense as well?

Yes, companies like Google and Microsoft already use AI to detect anomalies and predict attacks, but offensive use of technology is accelerating faster than defenses.

Has this attack ever affected companies in Brazil?

Google did not confirm specific victims in Brazil, but warns that global corporate systems were in sight. Brazilian companies using TOTP-based 2FA solutions should check updates urgently.

Read the original report on The Hacker News.