Interpol: AI-Powered Scammers Earn 4.5 Times More and Are Heading Toward Industrial Scale

The use of artificial intelligence is already visibly changing the economics of online fraud. Schemes in which criminals use AI bring them on average 4.5 times more profit, and Interpol warns that as AI agents develop, such operations may reach a nearly industrial level.

Why revenue is growing

The main effect of AI for scammers is not magic but a sharp drop in the cost, and a speedup, of familiar tasks. Where people once had to write dozens of email variants by hand, correspond with victims and adapt messages to language, tone and context, a model now does a significant part of the work. It generates text faster, adapts its style to a specific person and helps launch more attempts in the same amount of time. As a result, old schemes work more efficiently without a proportional rise in spending on people and infrastructure.

The 4.5-fold profit increase points not only to better-quality deception but to a new economy of scale. If a fraud campaign was once limited by the number of operators, preparation time and language barriers, AI removes several of these constraints at once. What is especially dangerous is that automation helps test wording quickly, change scenarios and reuse successful templates. For criminals this means more reach, higher conversion and less manual routine.

What agents change

The next risk involves not just generating text or voice but AI agents that can carry out chains of actions almost autonomously. Such a tool can not only write an email but also collect data, send out messages, hold a conversation following a script and adapt as the conversation unfolds. If these capabilities move into criminal scenarios, fraud will stop being a set of separate episodes and become a managed process with minimal human involvement at each step.

Interpol fears that as AI agents develop further, cybercrime may take

on "industrial scale."

This wording matters because it is not about isolated experiments but about the risk of a mass-production line. The cheaper each contact with a potential victim becomes, the more profitable it is for criminals to increase volume. Even if the share of successful attacks does not change radically, profit grows through speed, personalization and near-continuous operation.

How schemes are changing

AI does not necessarily invent new types of fraud. It strengthens known methods and makes them scalable. In practice, this means threats users already know become more convincing, faster and cheaper to launch. Especially vulnerable are scenarios that require processing data in bulk, quickly changing the cover story or running many parallel conversations.

• Personalized phishing emails and messages without obvious language errors
• Quick adaptation of text to the victim's job title, country or specific situation
• Automated correspondence and handling of typical objections
• Mass launch of new campaigns with low testing costs
• Reuse of successful scenarios with almost no manual rework

This raises the burden on both ordinary users and companies. Employees find it harder to tell a fake message from normal business communication, and security teams must respond not only to individual attacks but also to the high speed at which they are updated. If criminals get cheaper automation tools, defenders have to spend more resources on training, filtering and checking suspicious actions.

What this means

The 4.5-fold profit increase shows a simple thing: AI has become not only a productivity tool for business but also an amplifier of criminal schemes. The faster more autonomous agents reach the market, the more important it is to protect processes, practice digital hygiene and verify unusual requests. Otherwise scammers will gain a real economy of scale.