Artificial intelligence has become the center of the tech industry’s biggest ambitions. Companies promise it will transform healthcare, productivity, education and entertainment. But the same technology powering those breakthroughs is also giving cybercriminals new tools to deceive people at an unprecedented scale.

Google says it has now encountered exactly that scenario.

The tech giant has filed a lawsuit against a Chinese cybercrime operation accused of using Google’s Gemini AI platform to fuel a sprawling scam network that targeted hundreds of thousands of victims. According to the company, the alleged scheme involved fake websites, millions of fraudulent messages, and sophisticated impersonation tactics designed to trick people into handing over money and sensitive information.

The case marks Google’s first coordinated legal action of this kind against an AI-driven scam network, and executives say it likely won’t be the last.

Google Says Gemini Was Exploited to Power a Massive Fraud Scheme

Google alleges that a group known as Outsider Enterprise used Gemini to help scale an elaborate operation built around impersonation and deception.

According to the lawsuit, the organization allegedly generated fake websites designed to resemble trusted institutions and well-known brands. The targets reportedly included Google itself, YouTube, the United States Postal Service, and New York’s E-ZPass toll service.

The objective was simple but devastatingly effective: convince users they were interacting with legitimate services before directing them toward fraudulent payments or harvesting personal information.

Google has requested a restraining order against the network as part of its legal action, arguing that the operation represents a coordinated transnational criminal enterprise rather than isolated cyber incidents.

“This is our first coordinated effort and lawsuit and that speaks to the breadth of impact that this particular scam has,” Google’s general counsel DeLaine Prado told The New York Times.

While Google didn’t specify exactly what internal safeguards failed or what additional controls it implemented within Gemini after discovering the abuse, the company acknowledged that AI tools can be manipulated by bad actors seeking to industrialize fraud.

The lawsuit reflects a growing concern throughout the technology sector: the very systems designed to help people create and communicate can also dramatically amplify criminal activity.

The Numbers Behind the Scam Are Staggering

What makes this case particularly alarming is its scale.

Google says the alleged operation affected “hundreds of thousands of victims,” with financial losses reaching into the millions of dollars.

The figures paint a troubling picture of how quickly AI-assisted scams can spread.

According to the company, the network created approximately 9,000 fraudulent websites and generated one million fake URLs tied to the scheme.

Over just a two-week period, Android users flagged around 55,000 spam text messages connected to the operation. Google also identified approximately 2.5 million messages containing links directing users to fraudulent websites.

Those numbers come from a single criminal organization.

For cybersecurity experts, that’s perhaps the most concerning part.

Traditional phishing campaigns often relied on smaller teams manually creating scam materials. AI allows fraudsters to automate much of that work, producing convincing content at speeds that would have been unimaginable only a few years ago.

Instead of targeting dozens of people, criminals can now target millions simultaneously.

Google Teams Up With the FBI and Major Carriers

Google isn’t handling the situation alone.

The company said it coordinated its response with the FBI, as well as major telecommunications providers including AT&T, T-Mobile and Verizon.

The collaboration highlights how the fight against AI-powered scams increasingly requires multiple industries working together.

Phone carriers monitor suspicious messaging activity.

Technology companies identify malicious infrastructure.

Law enforcement investigates criminal networks operating across borders.

None of those entities can effectively address the problem independently.

“Criminals increasingly use AI to make fraud like this more convincing and harder to detect,” FBI Assistant Director Brett Leatherman said. “And we need a permanent solution to bring them to justice.”

The FBI’s comments underscore an uncomfortable reality.

Scams themselves aren’t new.

What’s changed is the sophistication.

AI-generated language can mimic trusted institutions with fewer grammatical mistakes and more natural communication patterns. Fraudulent websites can be created rapidly and tailored to specific audiences. Messages can evolve dynamically based on victim responses.

The result is a threat landscape becoming harder for ordinary consumers to navigate.

Why Google Is Pushing for New Laws

Google argues that the legal framework currently in place isn’t sufficient to deal with this new generation of scams.

As a result, the company is backing several bipartisan legislative proposals designed to strengthen protections against fraud.

Among the measures Google supports are the National Strategy for Combatting Scams Act, the Strategic Task Force on Scam Prevention Act, the STOP Scams Against Seniors Act, and the AI Plan Act.

The company’s position is clear: AI has fundamentally changed the scale and speed of digital fraud, and regulations need to evolve accordingly.

“This is not spam. It is organized transnational crime moving through our phones, and it demands a response as coordinated and aggressive as the threat itself,” Congressman Brian Fitzpatrick said.

That distinction matters.

For years, consumers were taught to think of scam messages as annoying inconveniences filling inboxes with obvious warning signs. Today’s scams increasingly resemble legitimate communications, making traditional advice less effective.

The challenge has shifted from spotting broken English and suspicious links to recognizing sophisticated impersonation campaigns built with cutting-edge tools.

AI’s Biggest Challenge May Be Trust

The irony of this situation isn’t lost on the industry.

Technology companies have spent the past several years promoting artificial intelligence as the future of digital experiences. Investors have poured billions into the sector, while consumers are rapidly integrating AI assistants into everyday life.

At the same time, the technology’s misuse threatens one of the internet’s most valuable currencies: trust.

If users begin questioning whether every text message, website notification, or customer service interaction could be part of an AI-driven scam, confidence in digital communication itself starts to erode.

Google’s lawsuit sends a message that companies can no longer treat these incidents as isolated cybersecurity problems.

They’re becoming business problems.

Policy problems.

Public safety problems.

The case against Outsider Enterprise may ultimately determine little on its own, particularly given the complexities of pursuing cross-border cybercriminal networks.

But it represents something larger.

The AI arms race isn’t only happening between tech giants competing to build smarter models. It’s also unfolding between defenders trying to protect users and criminals eager to exploit powerful new tools.

And if Google’s allegations are accurate, that battle has already entered a dangerous new phase.