Artificial intelligence is rewriting the playbook for crime, from cheap deepfake scams and AI-written ransomware to mass identity hijacks and critical-infrastructure hacks.
Why it matters: This new class of AI-supercharged crime is putting lives and financial systems at risk. But police training, laws and cross-border tools aren't keeping up, futurists tell Axios.
Off-the-shelf AI lowers the skill level and cost of carrying out attacks, enabling small crews to execute schemes that previously required nation-state resources.
Crimes can now hit millions at once with voice clones and account takeovers, while local agencies are trained and funded to chase one case at a time.
How it works: AI can create automations to "lock pick" into a system millions of times per second, something humans can't do, futurist Ian Khan tells Axios.
Once inside, hackers can then use AI to steal identities, pump and dump stocks and cause havoc to water plants, smart homes and hospitals.
The attacks can come from across the street to the other side of the world, said futurist Marc Goodman, author of "Future Crimes: Inside the Digital Underground and the Battle for Our Connected World."
Deep fake voices can convince victims to hand over money, or stolen identities could lead to voter fraud, child pornography, and false arrests.
The purpose: anything from extorting money to causing pain to millions.
The latest: Chinese state-backed hackers used AI tools from Anthropic to automate breaches of major companies and foreign governments during a September cyber campaign, the company said Thursday.
"We believe this is the first documented case of a large-scale cyberattack executed without substantial human intervention," the company said in a statement.
State of play: Deepfake fraud attempts surged 3,000% in 2023, per DeepStrike, a cybersecurity group.
U.S. losses from fraud that relies on generative AI are projected to reach $40 billion by 2027, according to the Deloitte Center for Financial Services.
Generative AI has increased the speed and scale of synthetic-identity fraud, especially across real-time payment rails, according to the Federal Reserve Bank of Boston.
A deepfake attack occurred every five minutes globally in 2024, while digital-document forgeries jumped 244% year-over-year, the Entrust Cybersecurity Institute found in a report.
Zoom in: Beyond large-scale attacks, even petty AI crimes have local law enforcement on edge.
Some AI-powered drones could collect data on the best places to bury bodies on less-traveled roads.
Future robo-dogs could burglarize homes.
Hacked cars may just drive off by themselves to chop shops, and AI systems could inform a would-be thief the best way to break into a car.
The bottom line: Few police academies are training cadets on spotting AI or computer crimes, both Khan and Goodman said.
That's leaving enforcement to federal authorities, who then need international cooperation to stop worldwide syndicates.
However, Miami Dade College this year announced it will be one of the first U.S. police academies to train cadets using a dedicated AI-assistant tool, in partnership with the AI company Truleo, the Miami New Times reports.
The Police Executive Research Forum (PERF) has been encouraging agencies to develop policies and training as AI policing tools become more prevalent.
You must be logged in to post a comment.