For years, one common way to spot phishing and other electronic fraud schemes was to look for poor spelling, awkward grammar, or suspicious formatting. That guidance is no longer enough. Generative artificial intelligence (GenAI) can help fraud perpetrators produce polished emails, realistic audio, and convincing video in seconds. As a result, phishing emails may now appear free of error, personalized, and consistent with messages your team receives every day.

The Federal Bureau of Investigation (FBI) has warned that criminals use GenAI to support financial fraud, including by improving translations, reducing spelling and grammar mistakes, and creating fraudulent documents or identities. In its 2025 Internet Crime Report, the FBI also reported that cyber-enabled crimes generated nearly $21 billion in reported losses, with cryptocurrency and artificial intelligence-related complaints among the costliest categories.

So, what can your business do to respond to fraud risks in the GenAI age?

Tools In Your Toolbox

Fortunately, potential phishing victims still have resources available. For example, instead of evaluating a message’s legitimacy based on its writing quality or appearance, you might question its intent and whether the request makes sense. No matter how professional an email or video appears, if it creates urgency, asks for secrecy, or requests login credentials, it should raise concerns. This is particularly true for payment-related emails.

For example, an email that appears to be from one of your business’ suppliers may seem valid. But if it asks you to respond immediately to prevent serious consequences (such as delivery delays of urgently needed materials), instructs you to keep the contents to yourself, or provides a link to a website that requires a login and password, be wary. In such cases, the prudent response is to call the supposed sender using a phone number you already have on file before replying to the email.

Seeing Isn’t Necessarily Believing

Apply the same caution to possible deepfakes, or AI-generated video and audio that look and sound realistic but are intended to deceive. Although you may be able to detect a deepfake by searching for video or audio glitches or other signs that it’s not real, it’s best to verify the sender rather than the media. So, if a “person” on the phone or in a video conference claims to be someone you know and asks for money, confidential information, or login credentials, verify the request directly with that person through another channel, ideally face-to-face.

Also be careful with potentially manipulated documents. Rather than relying on a single document to justify an action (such as authorizing a network user, changing a payer’s address, or “correcting” account numbers), compare the document with previously verified records. If you still doubt the request, ask the sender for additional documentation that would be difficult for unauthorized sources to create or access on short notice.

Adopt Useful Habits

Your organization may conduct cybersecurity awareness and other antifraud training programs to keep employees current on the latest scams. Unfortunately, given the pace of change, security training can quickly become outdated. Today, the most effective defense isn’t studying every new fraud method, which is impractical and time-consuming, but adopting certain habits. For example, employees should:

• Pause and think before responding to emails, texts, or phone requests
• Verify financial and other high-risk requests through one or more independent, trusted channels
• Escalate suspicious situations to a supervisor
• Report attempted or successful fraud schemes to the appropriate party (generally your IT department)

 
The bottom line: Employees should follow your business’ approval procedures for payment authorizations, account updates, and other requests.

Paradigm Shift

Red flags haven’t completely disappeared in the age of AI. But spotting them may require a little more scrutiny (sometimes with AI-enabled apps). Help your workers understand this paradigm shift and adjust their behaviors accordingly. We can help with internal controls and if you suspect fraud, an investigation.