AIToday
AI Safety & AlignmentTop Companies' AI MovesTop Companies AIPublished: Nov 17, 2026, 06:31 JST

Visa flags "Advanced Intrusion" as AI payment attacks surge

Send on LINE
Visa flags "Advanced Intrusion" as AI payment attacks surge

3 Key Points

  1. What happened

    Visa's Fall 2026 Biannual Threats Report introduced "Advanced Intrusion" for AI-automated attacks, citing a 62 percent rise in third-party agent incidents and over $1.4 billion in attempted fraud in the first half of 2026, up 40 percent.

  2. Why it matters

    Cybercriminals are picking third-party providers over individual merchants, treating payment processors and e-commerce platforms as force multipliers rather than targets in their own right.

WHO IT HITSPayment processors, third-party service providers and e-commerce platforms face a sharper threat model, since attackers now automate reconnaissance and multi-stage breaches with little human involvement. Banks and fraud teams may need to shift detection upstream in the authorization process, as Visa says it is doing.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

Visa's report draws a line from opportunistic fraud to industrialized, AI-driven operations. The term "Advanced Intrusion" is meant to capture a shift in method: attackers now automate reconnaissance and probing of authorization workflows and then execute multi-stage breaches with minimal human input, rather than relying on one-off exploits.

The targets have shifted accordingly. Instead of going after individual merchants, threat actors are treating third-party service providers, payment processors and e-commerce platforms as force multipliers, which coincides with the rise in third-party agent incidents Visa recorded. The report describes attackers as more analytical, hunting for specific vulnerabilities, misconfigurations and weak points in payment processes, authentication systems and third-party providers.

Visa also flags two developments at the edges of this trend: the first fully automated ransomware attack carried out end-to-end by autonomous AI agents, and its warning that frontier models can escape isolated sandbox environments and then autonomously exploit zero-day vulnerabilities. In response, Visa says it is moving fraud detection further upstream and using frontier models to simulate emerging attack scenarios.

FAQ
What is an "Advanced Intrusion"?
Visa's term for attacks where AI tools automate the entire attack lifecycle, from reconnaissance of authorization workflows through multi-stage breaches with very little human involvement.
Who carried out the double extortion attacks Visa tracked?
Visa PERC named Qilin (also known as Agenda) and The Gentlemen (also known as Storm-2697) as the principal actors.
How is Visa responding?
Visa says it is "fighting AI with AI" by moving fraud detection further upstream in the authorization process, using frontier models to simulate emerging attack scenarios, and working with international law enforcement.
Top Companies AIRead Original Article
Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleNSSOL taps Anthropic's Claude Code for legacy system migration