How AI is Changing Cybercrime
Kalamazoo, MI | October 14, 2026
Imagine receiving a phone call from your boss asking you to approve a payment.
The voice sounds authentic, the request makes sense, and the details are correct.
But there's a problem: the person on the other end isn't your boss. It's an AI-generated imitation.
It may seem like science fiction, but scenarios like this are already happening around the world. Cybercriminals are using AI to create convincing messages, impersonate trusted individuals, and scale attacks that once required considerably more time and expertise.
But here’s the thing: AI isn’t inventing new cybercrime. Instead, it’s just empowering the cybercrime that already exists.
So, without further ado, here are three major ways AI is changing the cyber threat landscape.
1. Ransomware: Fueling a More Efficient Criminal Ecosystem
Ransomware is malicious software that encrypts files or disrupts access to systems, with the goal of making the victim pay a ransom to get their files back.
And while the fundamental principles of ransomware haven’t changed, AI has launched ransomware into a new era.
With the help of AI, cybercriminals can perform the same functions with much faster speed:
Researching potential victims
Crafting initial-access messages
Assisting with technical tasks.
That said, AI’s role is still indirect: criminals still need access to a victim's systems, and AI alone can’t create sophisticated malware (yet).
A real-world example is RaccoonO365, a subscription-based phishing service disrupted by Microsoft in September 2025. Its operators supplied criminals with tools that impersonated Microsoft login pages and harvested account credentials.
Microsoft reported that the service had been used to steal at least 5,000 credentials across 94 countries. In January 2026, Microsoft explained how services like RaccoonO365 can facilitate ransomware and other follow-on crimes by providing attackers with an initial foothold.
The takeaway is that ransomware defense begins before encryption starts. Organizations should secure accounts, patch vulnerable systems, maintain offline or otherwise protected backups, and restrict access to sensitive resources.
If someone claiming to be your boss, a relative, or a financial institution asks you to transfer money or reveal sensitive information, verify the request through a separate, trusted channel. At work, require independent approval for significant payments and changes to vendor banking details. Most importantly, don't let urgency override common sense.
2. Social Engineering: Making Impersonation More Convincing
Buckle in, because this one’s freaky.
Social engineering exploits human trust rather than relying solely on technical vulnerabilities.
By posing as colleagues, relatives, or service providers, attackers can persuade victims to transfer money, disclose confidential information, or even grant access to accounts.
Generative AI makes impersonation easier by helping criminals imitate writing styles, produce realistic voices, and construct convincing supporting details. In fact, deep-fake audio and video can further blur the line between genuine communication and fabricated evidence.
You read that right: Criminals can use AI to convincingly pose as people you know in real life.
In May 2025, the FBI warned of a campaign in which malicious actors impersonated senior U.S. officials through text messages and AI-generated voice messages.
The attackers attempted to establish rapport with targets before directing them toward malicious links designed to obtain account access.
The lesson is simple: familiarity is no longer reliable proof of identity.
If someone requests an unusual payment, asks for confidential information, or urges you to act urgently, verify the request through a separate, trusted channel.
For businesses, independent payment approvals and established verification procedures can prevent costly mistakes.
3. Phishing: Personalized Scams at Greater Scale
For those unaware: phishing is a scam that tricks people into clicking malicious links that harm their computer.
Traditional phishing messages often contained obvious grammatical errors or awkward phrasing. So with a little education, most people learn to catch it pretty fast.
But AI has changed things.
Attackers use AI to adapt messages to different audiences, generate convincing business correspondence, and produce variations of the same scam at scale.
That means a phishing email may look professionally written, refer to a legitimate project, and appear tailored specifically to its recipient.
Microsoft has documented threat actors using large language models to support social engineering, including drafting phishing messages and preparing content for targeted campaigns.
In a separate example, the FBI warned in May 2026 about Kali365, a phishing-as-a-service platform offering AI-generated lures and tools to steal Microsoft 365 authentication tokens. Such tokens can enable account access even when conventional passwords are not enough to authenticate.
Because polished language is no longer a reliable indicator of legitimacy, users should inspect links carefully, access important services through known websites, and enable phishing-resistant authentication where possible.
Staying Ahead of AI-Powered Cyberattacks
AI is changing cybercrime in ways that we never saw coming.
But even though it’s frightening, there are ways that you can keep yourself safe online—even when facing an AI threat. But more on that in our next blog post.
For more information on cybersecurity and AI for your business, click here to contact us today!
