🤖Generative AI changed the scale
The FBI created a dedicated descriptor for "AI-related" complaints in 2026 — 22,000 reports and nearly $900 million in losses in the very first year. On the voice side alone, deepfake fraud attempts rose 1,300% in 2024. AI doesn't create new scams: it lets a single scammer run hundreds at once, with unprecedented realism.
FBI (IC3, 2026) · Pindrop, 2025 Voice Intelligence and Security Report🗂️Data breaches fuel the targeting
The average global cost of a data breach reached $4.44 million in 2025 ($10.22 million in the United States, a record). Every large-scale breach puts names, phone numbers and addresses into circulation — the raw material for a scammer who personalizes their approach rather than calling at random.
IBM, Cost of a Data Breach Report 2025📡A structured industry behind spam
The European operation "SIMCARTEL" (October 2025) exposed a single SIM-farm network that made 49 million fake online accounts possible. This isn't a cottage industry anymore: it's an on-demand rental service, operating at industrial scale.
Europol & Eurojust, October 2025📞Raw volume keeps climbing
The average monthly volume of fraud and telemarketing calls rose from 2.14 billion in 2024 to 2.56 billion per month in 2025 — up 20% in a year. Losses from fraudulent calls specifically targeting people over 60 topped $4.57 billion in 2024, in the United States alone.
PIRG, Ringing in Our Fears 2025 · FBI (IC3), 2024💻Remote work blurs the usual cues
With a growing share of the workforce fully or partly remote, professional exchanges by text, video call or messaging app have become the norm — which makes a fake job offer or a fake message from a "colleague" far more credible than it was a decade ago, when those exchanges went almost exclusively through a verifiable work email.
See also the recruitment fraud reportWhat these five factors have in common
None of these changes is isolated — they reinforce each other. A data breach supplies the raw material, a SIM farm supplies the number, generative AI supplies the script and sometimes the voice, and remote work supplies a context where receiving an unexpected work message no longer surprises anyone. It's this combination, not a single factor, that explains the acceleration.
Frequently asked questions
When did scams really start exploding?
The main indicators (losses reported to the FBI, fraudulent-call volume, cost of data breaches) show continuous acceleration over several years, with a particularly sharp jump since the arrival of mainstream generative AI.
Is artificial intelligence the only factor?
No. It adds to other already-documented factors: the explosion of data breaches that feed scammers personal information, the spread of remote work, and a structured industry (SIM farms, call centers) that has professionalized fraud at scale.
Is this going to stop?
Nothing in current reports suggests a slowdown — quite the opposite, several bodies (Europol, ENISA) document growing sophistication. That's exactly why protection that keeps adapting continuously, rather than a fixed list, remains necessary.
Other markets we cover
The indicators above are American, European or global — but the same combination of drivers (AI, breached data, call volume) shows up just as clearly in national numbers elsewhere.
🇦🇺Australia — first-half reports jumped 26% in a single year
The National Anti-Scam Centre logged $2.18 billion in reported scam losses across 2025, up 7.8% on 2024. The acceleration shows most clearly mid-year: the $173.8 million lost in the first half of 2025 was 26% higher than the same period in 2024.
National Anti-Scam Centre (ACCC), Targeting Scams report, March 2026.🇬🇧United Kingdom — a record 444,000+ fraud cases in 2025
Cifas, the UK's national fraud database, recorded more than 444,000 fraud cases in 2025 — the highest number in a single year since records began, a 6% rise on 2024. Investigators partly attribute the increase to fraud shifting toward channels, like messaging apps and AI-generated content, that are harder to categorize under older reporting systems.
Cifas, National Fraud Database, 2025; reported by digit.fyi.🇨🇦Canada — a record $704 million lost, the worst year on record
The Canadian Anti-Fraud Centre logged over 112,000 reports and more than $704 million in losses in 2025 — up from $643 million in 2024, itself a record at the time. Investment fraud alone, often built around fabricated trading dashboards, accounted for $351 million of the total.
Canadian Anti-Fraud Centre, 2025 annual results.🇮🇳India — cyber-fraud losses jumped 206% in a single year
Indians lost over ₹22,845 crore to cyber fraud in 2024, a 206% increase from 2023, according to data the Ministry of Home Affairs shared in Parliament. The National Cybercrime Reporting Portal had already logged 1.25 million incidents by the end of June 2025 alone — nearly five times its 2021 volume.
Ministry of Home Affairs (India), data shared in Parliament, 2025; I4C, National Cybercrime Reporting Portal.Go further
Egidio — The Threat Laboratory, "Why fraud is exploding right now", egidio.app/en/laboratoire/why-now/. Licensed CC BY 4.0.
Free to reuse, including commercially, with attribution. Reuse terms.