Criminals are using artificial intelligence to sharpen old tricks, and the losses are adding up fast. In the United States last year, people lost more than $893 million to crimes that used AI, according to the FBI. The figure signals a growing threat that touches families, workers, and businesses across the country.
The warning comes as AI tools become cheaper and easier to use. Fraudsters now mimic voices, write convincing messages, and create fake images with little effort. The new tools speed up scams, and they can make lies look and sound real.
“Americans lost more than $893 million from AI-related crimes last year, according to the FBI.”
How AI is supercharging old scams
AI has not invented fraud. It has made fraud faster and harder to spot. Scammers use text generators to write perfect emails that impersonate banks, bosses, or family members. They use voice cloning to copy a person’s speech from short clips on social media. They can even produce deepfake video to stage fake meetings or change an identity in real time.
These tools lower the skill level needed to run a con. A single person can send thousands of customized messages. One fake audio clip can start a panic. A short video can trick a team into wiring money or sharing logins.
Who is at risk and why it works
Anyone who answers a phone, opens an email, or uses social media can be a target. The victims include students, retirees, small firms, and large companies. Pressure and familiarity are the weapons. A cloned voice that sounds like a child or a manager can push someone to act before they think.
AI helps criminals remove the usual warning signs. Bad spelling is gone. Odd phrasing is fixed. The caller ID and profile photo may look right. People trust what they see and hear, so they move quickly. That is exactly what the scammers want.
Inside the number
The $893 million figure points to a surge in AI-enabled crime, but it may undercount the harm. Many victims do not report. Some never learn that AI was used. Others feel embarrassed or believe police cannot help. The FBI has urged the public and businesses to share reports because data can reveal patterns and stop repeat attacks.
Experts say the rise of AI-driven fraud tracks with the spread of easy-to-use apps. As more tools hit the market, criminals test them in phishing, investment schemes, romance scams, and fake tech support. The speed of change makes defenses harder to plan if organizations do not update training and controls.
What authorities and platforms are doing
Law enforcement agencies have increased outreach and training on AI threats. They promote reporting portals and work with industry to remove illegal content. Some platforms now label synthetic media or limit tools that can be misused. Cyber insurance carriers also ask clients to add checks for payments and identity verification.
These steps help, but the gap remains. Detection tools miss some fakes. Labels can be stripped. Criminals move across borders and switch tactics. That is why experts stress layered defenses and steady education.
How to protect yourself right now
- Set a family and workplace “safe word” for urgent money or data requests.
- Use a second channel to verify unexpected calls, texts, or emails.
- Slow down. Scammers push for speed and secrecy.
- Keep accounts private. Limit public audio and video that could feed voice cloning.
- Enable multi-factor authentication on email, banking, and cloud tools.
- Train teams to spot deepfake tells, like odd lighting, lip sync issues, or audio glitches.
- Report incidents to the FBI and your bank as soon as possible.
What to watch next
AI will keep getting better at sounding and looking real. That means more convincing messages and calls. It also means a race to build stronger detection and clear rules for synthetic media. Banks and employers are adding call-back rules and payment holds for risky transfers. Schools and community groups are teaching people how to pause and check.
The latest figure from the FBI shows the cost of inaction. The scams are faster, and the targets are wider. Strong habits, clear verification steps, and quick reporting can blunt the damage. Readers should expect more warnings, more tools to label fakes, and more cases that test new laws.
The bottom line is simple. Treat urgent digital requests with caution, even if the voice or video looks perfect. Trust, then verify. The minutes you take to confirm could save thousands of dollars.