AI Voice Scam Costs Intesa Sanpaolo €95 Million

AI voice cloning and a fake WhatsApp message drained €95 million from Intesa's private bank, exposing the limits of executive verification.

Last Updated: September 25, 2026 Editorial Process
Editorial Process
See more of Inside AI's trusted news by adding us as a preferred source on Google.
AI neural network visualization
Published on: September 25, 2026

September 25, 2026, (Inside AI) — Fraudsters armed with artificial intelligence impersonated top executives at Intesa Sanpaolo, Italy’s largest bank, and siphoned €95 million ($108 million) from its private banking unit Fideuram, according to two people familiar with the matter. More than half the money was clawed back, but €36 million remains missing after being routed through a web of overseas accounts and converted into cryptocurrency.

The scheme, which began in February, exploited a simple but devastating weakness: human trust in familiar voices and names. It also marks one of the largest known AI-assisted corporate thefts, raising urgent questions about how banks verify executive instructions in an era of cheap, convincing deepfakes.

Then-Fideuram Chairman Paolo Molesini received what appeared to be a WhatsApp message from Intesa CEO Carlo Messina, asking for urgent help with an overseas transaction. A follow-up phone call seemed to come from a senior partner at a prominent law firm, confirming the request. The callers used AI to replicate the lawyer’s voice, the sources said. Believing the instruction was genuine, Molesini told his finance department to arrange transfers to foreign accounts, mainly in China and Hong Kong.

Intesa Sanpaolo and Fideuram declined to comment. Molesini did not respond to requests for comment. He resigned as chairman in March, citing personal reasons, and Fideuram gave no further explanation at the time.

Read: AI Scam Uses Holiday Photos to Pinpoint Travelers and Steal Bank Details

Fideuram’s internal controls eventually flagged the transfers as irregular, prompting alerts to banks and authorities in multiple countries. Roughly €53 million was recovered through cooperation between officials in China, Portugal, and Italy, according to the sources. The remaining funds have not been traced.

Milan prosecutors have placed a foreign national living outside Europe under investigation on suspicion of computer fraud, one source said. Neither Molesini nor any other Fideuram executives are under investigation.

The incident echoes a 2025 case in which fraudsters used AI to mimic the voice of an Italian minister and persuaded businessman Massimo Moratti to transfer nearly €1 million to an overseas account. That money was later recovered. The Fideuram case is far larger and highlights how quickly AI voice cloning has moved from novelty to a tool for high-value corporate crime.

Security researchers have warned for years that voice authentication and executive approval workflows are vulnerable to synthetic media. A 2024 study by McAfee found that a majority of people could not reliably distinguish between a real voice and an AI-cloned one after just a few seconds of audio. Banks have since invested in callback verification and multi-factor approvals, but attackers continue to adapt.

“The problem is not the technology itself, but the speed at which it can be deployed against human psychology,” said Dr. Elena Bertolini, a cybersecurity researcher at Politecnico di Milano, who was not involved in the case. “When a request appears to come from a trusted CEO and is confirmed by a second familiar voice, even experienced executives can be fooled.”

The recovery of €53 million involved coordination across three jurisdictions, a process that took weeks. The missing €36 million likely moved through shell accounts and crypto exchanges, making tracing difficult. Authorities have not disclosed the identity of the foreign national under investigation.

Read: Russian Hacker Used Google Gemini AI for 89% of Cybercrime Operations

For Intesa Sanpaolo, the financial hit is manageable. The bank reported net income of €8.5 billion in 2025, so €36 million represents less than half a percent of annual profit. But the reputational damage and the precedent are more significant. Clients of private banks expect discretion and security. A breach that relied on impersonating the CEO and a lawyer could undermine confidence in Fideuram’s controls.

Italian regulators have not publicly announced new rules in response. However, the case is likely to accelerate discussions in the European Union about corporate verification standards under the AI Act, which began phasing in obligations for high-risk AI systems earlier this year. Voice cloning tools are widely available, often free, and require only a few seconds of sample audio.

Fideuram has not said whether it has changed its internal approval procedures. The bank’s parent, Intesa Sanpaolo, operates in 12 countries and manages over €1 trillion in assets. The incident is a reminder that even the most sophisticated institutions can be breached through a single trusted channel.

As Milan prosecutors continue their investigation, the missing millions may never be recovered. The case stands as a warning: in the age of AI, a phone call is no longer proof of identity.

More from Inside AI

  • AI Tools

    Google Photos Expands AI Wardrobe Tool to US, Brazil, India

    September 26, 2026
  • AI In Business

    Context Engineering Overtakes Prompting As AI Leaders’ Key Skill

    September 26, 2026
  • Vibe Coding

    Anthropic Outlines Six-Step Framework for AI-Driven Code Modernization

    September 26, 2026
  • AI Safety

    OpenAI Agent Breached Australian Medicare Portal, Pattern of Rogue AI Activity Widens

    September 26, 2026
  • Agentic AI

    Microsoft Gives AI Agents Their Own Email, Calendars in Biggest Copilot Update

    September 26, 2026
  • AI Tools

    AWS Adds AI Agent Skills to MCP Server for SES and End User Messaging

    September 26, 2026
  • AI Policy & Regulation

    FTC Chair Andrew Ferguson Rejects Treating AI Agents as Independent Actors

    September 26, 2026
  • AI Policy & Regulation

    Perplexity AI Sued by Startup DaVoice Over Alleged Wake Word Trade Secret Theft

    September 26, 2026

Never Miss a Breakthrough

Join 50,000+ readers who get our daily AI intelligence briefing. No fluff, just what matters.

Join Our Newsletter Community

Subscribe

Inside AI is an independent publication covering artificial intelligence news, machine learning research, and the tools shaping the future of technology. No hype. Just what's happening in the AI world.

Topics

  • Artificial Intelligence
  • Machine Learning
  • Generative AI
  • Agentic AI
  • Vibe Coding
  • Prompt Engineering
  • AI Policy & Regulation
  • AI Hardware & Infrastructure
  • AI Tools
  • AI In Business
  • Robotics
  • Cybersecurity AI
  • AI Safety
  • AI Tools & Reviews (Coming soon)

Company

  • Editorial Standards
  • Privacy Policy
  • Terms of Service
  • Contact
  • About Us

Others

  • Press Releases
  • Features
  • Sponsored Content
  • Advertise with us
  • Newsletter

© 2026 Inside AI. All rights reserved.

Designed by Blue Flare Digital