AI Phishing and Deepfake Scams: How to Protect Your Business

Phishing used to be easy to spot: bad spelling, strange grammar and a greeting like "Dear Customer." Those days are over. Criminals now use the same AI tools your team uses to write emails, and they're using them to write perfect, personalized scams in seconds. Some go further and clone a manager's voice or face. Here's what has changed, the warning signs to watch for, and the simple controls that stop these attacks cold.

What AI changed for attackers

AI didn't invent new crimes. It made old ones faster, cheaper and far more convincing:

  • Flawless, tailored emails. An attacker can feed an AI tool your website, your LinkedIn posts and a few public details, and get back an email that sounds exactly like a vendor, a client or your own boss. No typos, no awkward phrasing.
  • Scams at scale. What once took a criminal an hour now takes seconds, so they can send thousands of personalized messages instead of one generic blast.
  • Voice cloning. A few seconds of audio from a video, a podcast or a voicemail greeting is enough to imitate someone's voice on a phone call.
  • Deepfake video. In one widely reported 2024 case, a finance employee at a multinational company transferred roughly $25 million after a video call in which every other "colleague," including the CFO, was a deepfake.

Small businesses are a favorite target because everyone knows everyone, approvals are informal, and one person often handles payments end to end.

The scams we see most often

  1. The urgent payment request. An email or call that appears to come from the owner asks for a wire transfer, gift cards or a quick payment "before end of day."
  2. The changed bank details. A real-looking message from a vendor says their bank account has changed and the next invoice should go to a new one.
  3. The fake sign-in page. A shared document, voicemail or invoice link leads to a perfect copy of the Microsoft 365 sign-in page that steals passwords and MFA codes.
  4. The help desk impersonation. A caller posing as IT support asks an employee to approve a sign-in request or install a remote access tool.

Warning signs that still give scams away

Good grammar no longer means a message is safe. Instead, teach your team to watch for the situation, not the spelling:

  • Urgency and secrecy. "Do this now" and "keep this between us" are the two biggest red flags.
  • A change to how money moves. New bank details, a new payment method or a request to skip the usual approval.
  • An unusual channel. The boss suddenly texting from a personal number, or a vendor calling instead of emailing.
  • Pressure to approve something you didn't start, like an MFA prompt or a password reset.
  • Small glitches on calls, such as odd pauses, flat emotion or a refusal to turn the camera on or answer a personal question.

Five controls that stop AI scams

The best defenses aren't high-tech. They're simple rules that make a perfect fake useless:

  1. Verify using a number you already have. Any request involving money, bank details or passwords gets confirmed by calling the person back on a number from your records, never one in the message.
  2. Require two people for payments. Wire transfers and bank detail changes need a second approver. A scammer would have to fool two people through two separate channels.
  3. Agree on a code word. Owners and finance staff can share a simple phrase to confirm unusual requests by phone, which no voice clone will know.
  4. Use phishing-resistant sign-ins. Passkeys and stronger MFA stop stolen passwords and codes from working on fake sites.
  5. Make it safe to slow down. Tell your team, in writing, that no one will ever be in trouble for pausing to verify a request, even one that appears to come from the owner.

The technology that helps

Process stops the scams that get through, and technology reduces how many arrive in the first place:

  • Advanced email filtering catches impersonation attempts, look-alike domains and malicious links.
  • Email authentication (SPF, DKIM and DMARC) makes it much harder for criminals to send email that appears to come from your domain.
  • Security awareness training with realistic, AI-era phishing simulations keeps the warning signs fresh.
  • 24/7 monitoring spots suspicious sign-ins, new inbox rules and unusual activity if an account is compromised.

If you think you've been targeted

Speed matters. Don't reply to the message or call back the number it provides. If money was sent, call your bank immediately and ask them to recall the transfer. If a password was entered on a suspicious page, change it and contact your IT provider so the account can be checked for unauthorized access, forwarding rules or other changes.

How Harvest MSP can help

We help small businesses put these protections in place without slowing anyone down. That includes email filtering and domain authentication, phishing-resistant MFA, staff training and simulated phishing through our security services, and around-the-clock detection through our 24/7 security operations center. We can also help you write a simple payment verification policy your whole team can follow.

Want to learn how managed IT services can support your business? Email info@harvestmsp.com or call 630-634-5760.

← Back to all articles

Ready for next-level IT support?

Tell us about your business and we'll recommend the services that fit.
We pride ourselves on delivering honest solutions that fit your needs.