AI Abuse Statistics in US 2026 | Misuse, Fraud, Hacking, Cybercrime & Key Facts

In 2025, the FBI’s Internet Crime Complaint Center tracked AI-related fraud as a distinct category for the first time, tying it to 22,364 complaints and $893 million in reported losses out of a record $20.877 billion in total US cybercrime losses. Older Americans alone lost $352 million to AI-enabled scams, while the FTC separately reported $3.5 billion in imposter scam losses nationwide, both driven heavily by increasingly convincing voice cloning and deepfake technology.

AI Abuse in America 2026 – Introduction

AI abuse crossed a genuine threshold in 2025, moving from an emerging worry that security researchers debated at conferences into a line item that federal agencies now track and report with real dollar figures attached. For the first time in its 25-year history, the FBI’s Internet Crime Complaint Center broke out AI-related fraud as its own category, and the numbers behind that new category tell a clear story: criminals have adopted generative AI faster than most victims, banks, or even law enforcement have adapted to detect it. Voice cloning that once required a Hollywood studio budget now needs only a few seconds of audio pulled from a social media video, and phishing emails that used to be riddled with grammatical tells now read as fluently as a message from a real colleague.

What makes 2026 different from the AI-scam warnings of prior years is the scale of confirmed loss data now available. Total US cybercrime losses crossed $20 billion for the first time in 2025, a 26% jump in a single year, and agencies investigating that surge increasingly point to AI as an accelerant across nearly every major fraud category, from romance scams to business email compromise to elder fraud. Understanding exactly where AI is showing up in these crimes, and how much money it has already cost Americans, is essential context for anyone trying to make sense of how fast the threat landscape has shifted.

Interesting Facts About AI Abuse in US 2026

Data Source: FBI Internet Crime Complaint Center 2025 Annual Report; Federal Trade Commission Consumer Sentinel Network

These figures capture a fraud landscape being reshaped by two forces at once: more crime overall, and a growing share of that crime carrying an AI fingerprint. The jump to $20.877 billion in total losses and 1,008,597 complaints both represent all-time records for the IC3, and the $893 million tied specifically to AI in its debut reporting year almost certainly understates the true figure, since the FBI itself has acknowledged that most victims never realize AI was involved in the attack that targeted them.

The generational divide in this data is stark. Older Americans lost $352 million to AI-enabled scams specifically, and separately lost $7.7 billion to fraud overall, a 37% jump from 2024, with average losses per senior victim running close to double those of younger victims. That gap reflects both a targeting strategy, since scammers know older Americans often hold more accumulated savings, and a technology gap, since voice cloning and deepfake video are specifically effective against people less familiar with how convincingly AI can now fabricate a trusted voice or face.

AI Abuse & Cybercrime Losses in US 2026 | Total Scale & Growth

US CYBERCRIME LOSSES BY YEAR (billions)
2023        ████████████████████████                  $12.5
2024        ████████████████████████████████           $16.6
2025        ████████████████████████████████████████   $20.9

Data Source: FBI Internet Crime Complaint Center Annual Reports, 2023-2025

AI abuse is scaling inside a broader cybercrime environment that was already breaking records before AI entered the picture as a distinct tracking category. Total reported losses climbed from $12.5 billion in 2023 to $16.6 billion in 2024 to $20.877 billion in 2025, a 26% single-year jump that pushed the average loss per complaint to roughly $20,699. Complaint volume crossed one million in a single year for the first time in IC3’s 25-year history, with 452,868 of those complaints, or 85% of all reported dollar losses, falling into the broad “cyber-enabled fraud” category that increasingly overlaps with AI-assisted tactics.

The FBI’s decision to finally give AI its own reporting category in the 2025 report reflects how mainstream the technology has become inside criminal operations. 22,364 complaints referenced AI specifically, tied to $893,346,472 in reported losses, spanning voice cloning, deepfake video and images, and AI-generated scripted conversations used to sustain scams over multiple contacts with a victim. FBI Cyber Division officials have been explicit that this figure represents a floor rather than a ceiling, since attribution requires either the victim or the investigator to specifically recognize and flag AI involvement, something that happens inconsistently even when AI clearly played a role.

AI-Enabled Fraud Statistics 2026 | Investment Scams & Business Email Compromise

LARGEST US CYBERCRIME LOSS CATEGORIES, 2025 (billions)
Investment Fraud              ████████████████████████████ $8.65
Cyber-Enabled Fraud (crypto)  ██████████████████████████████████████ $11.0+
Business Email Compromise     ███████████████      $3.05
Tech/Customer Support Scams   █████     $1.23

Data Source: FBI Internet Crime Complaint Center 2025 Annual Report

Investment fraud remains the single most expensive category of cybercrime in America, with reported losses reaching $8.65 billion in 2025 across roughly 73,000 complaints, a category investigators increasingly link to AI-generated fake trading platforms, AI-written promotional content, and AI-cloned videos of trusted financial figures endorsing fraudulent schemes. Cryptocurrency-related complaints ran even higher in total dollar terms, at more than $11 billion across 181,565 complaints, reflecting how AI tools have made it easier for scammers to build convincing fake exchanges and personalized outreach at scale.

Business Email Compromise, where a scammer impersonates an executive or vendor to trick an employee into wiring funds, grew to $3.046 billion in losses, up from $2.77 billion the year before, a rise security researchers attribute in part to AI-generated emails that no longer contain the awkward phrasing that once made BEC attempts easier to spot. Phishing losses nearly tripled, from $70 million to $215.8 million, even as the raw number of phishing complaints held roughly flat, meaning each successful phishing attempt is now extracting significantly more money per victim than it used to, consistent with AI helping criminals craft more targeted, higher-value lures rather than simply more numerous ones.

AI Voice Cloning & Deepfake Scams 2026 | How Common They’ve Become

VOICE CLONING TECHNICAL CAPABILITY — RESEARCH FINDINGS
Accuracy from 3-Second Clip       ███████████████████████████████████ 85%
Accuracy With Extended Training   ████████████████████████████████████ 95%
People Unable to Detect a Clone   ███████████████████████████████████ 70%

Data Source: McAfee Labs Artificial Imposters Voice Cloning Research; Global Deepfake Fraud Industry Reporting

The technical barrier to convincing voice cloning has effectively collapsed. Research testing shows AI tools can now reproduce a person’s voice with 85% accuracy from just a three-second audio sample, rising to 95% accuracy with more extensive training material, and in surveys of thousands of respondents, roughly 70% of people could not reliably tell a cloned voice apart from the real thing. That combination means a scammer needs nothing more exotic than a public social media video or a voicemail greeting to generate a voice convincing enough to fool a family member during a stressful, time-pressured phone call.

Real cases already illustrate the pattern nationally: in one documented incident, a Brooklyn woman received a call that sounded exactly like her in-laws, followed immediately by a stranger claiming the couple was being held for ransom, when in fact the “in-laws’ voices” had been entirely AI-generated. Globally, deepfake-enabled fraud topped $200 million in losses in the first quarter of 2025 alone, and AI-generated phishing emails now achieve click-through rates roughly four times higher than traditional, human-written phishing attempts, a gap security researchers attribute to AI’s ability to eliminate the grammatical errors and generic phrasing that used to help people spot a scam on sight.

AI Scams Targeting Older Americans 2026 | Elder Fraud & Grandparent Scams

ELDER FRAUD LOSSES BY YEAR (billions)
2023        ██████████████████████████████████  $3.4
2024        █████████████████████████████████████████████ $5.6
2025        ██████████████████████████████████████████████████████████ $7.7

Data Source: FBI Internet Crime Complaint Center Elder Fraud Report 2025; Federal Trade Commission

Older Americans absorbed a disproportionate share of the damage from AI-enabled fraud in 2025. Total fraud losses among adults 60 and older reached $7.7 billion, a 37% increase over 2024, and within that total, the FBI specifically attributed $352 million across more than 3,100 victims to AI-related tactics, primarily voice cloning used in so-called “grandparent scams” or “distress scams,” where a fabricated emergency call convinces a senior their grandchild urgently needs money wired or transferred. Average losses per senior victim reached roughly $38,500, nearly double the average loss reported by younger filers.

The mechanics of the modern grandparent scam differ sharply from its decades-old predecessor. Where scammers once relied entirely on a victim’s imagination to fill in an unfamiliar voice on the phone, AI voice cloning now reproduces the actual sound of a specific grandchild, pulled from social media videos or voicemail greetings, making the emotional manipulation dramatically harder to resist even for people who have been explicitly warned about the scam. Reported losses tied specifically to distress and grandparent-style scams topped $5 million in 2025, a figure officials caution reflects only the fraction of cases victims chose to report at all.

AI-Generated Phishing & Hacking Statistics 2026 | Cybercrime on the Rise

PHISHING TRENDS — AI ADOPTION AND LOSS GROWTH
Phishing Emails Using AI (late 2024-early 2025)   ███████████████████████ 80%+
Phishing Loss Growth (2024 to 2025)               ███████████████████████ 3x

Data Source: ENISA Threat Landscape Analysis; World Economic Forum Global Cybersecurity Outlook 2026

The mechanics of AI-assisted hacking and phishing follow a consistent pattern across nearly every dataset available: AI does not typically introduce entirely new attack methods so much as it removes the friction that used to limit how many convincing attacks a single criminal operation could run at once. European regulators found that more than 80% of phishing emails analyzed in late 2024 and early 2025 used AI to some degree, whether to draft the lure itself, translate it into a target’s native language, or personalize it using scraped social media details, a shift that helps explain why phishing losses roughly tripled even as raw complaint counts stayed flat.

Corporate security teams have noticed the shift too. The World Economic Forum’s 2026 Global Cybersecurity Outlook found that 73% of organizations were directly affected by cyber-enabled fraud in 2025, and 87% now identify AI-related vulnerabilities as their fastest-growing cyber risk, surpassing both ransomware and supply chain attacks as the top concern among surveyed executives. Anyone looking for a broader picture of how consumer-facing fraud connects to this same underlying trend can find additional detail in Consumer Fraud Statistics in US, which tracks the full range of scam types reported to the FTC beyond the AI-specific slice covered here.

AI Fraud & Imposter Scams on Social Media 2026 | Impersonation Statistics

FTC IMPOSTER SCAM & SOCIAL MEDIA FRAUD — 2025 (billions)
Total Imposter Scam Losses           ███████████████████████████████ $3.5
Social Media-Originated Losses       ███████████████████████  $2.1
Business/Government Impersonation    ██████████████   $2.0

Data Source: Federal Trade Commission Consumer Sentinel Network 2025 Data

Impersonation has become the preferred entry point for scammers using AI tools, and the FTC’s 2025 data shows just how much money that shift has moved. Imposter scams cost Americans $3.5 billion, nearly triple the figure from 2020, with the largest losses coming from scammers posing as banks urging victims to “protect” their accounts by transferring funds, and business or government impersonators separately accounting for close to $2 billion in losses. A 40% jump in government imposter scam reports was driven substantially by fake toll-payment text messages, a scheme that spread nationally in 2025 using AI-generated, region-specific messaging that mimicked real toll agency communications.

Social media has emerged as the costliest single contact method for fraud, generating $2.1 billion in reported losses, an eightfold increase over five years, with roughly 30% of scam victims reporting their fraud began on a social platform. That growth tracks closely with AI’s ability to scrape public profiles, photos, and life updates to build far more convincing, personalized impersonation attempts than mass-market scam messages of a decade ago, whether that means a fake romantic interest built from a stolen photo set or a cloned voice pulled directly from a public video post. For context on how AI’s broader economic footprint is fueling both the tools behind these scams and the defenses against them, see AI Investment Statistics in US.

AI Abuse Law Enforcement Response 2026 | FBI, FTC & Recovery Efforts

FBI RECOVERY INITIATIVES — KEY RESULTS
Operation Level Up: Victims Notified            ██████████████████████ 8,000+
Operation Level Up: Losses Prevented ($M)       ██████████████████████ 500+

Data Source: FBI Internet Crime Complaint Center; Federal Trade Commission Report to Congress, December 2025

Federal agencies have started building dedicated response programs aimed specifically at AI-accelerated fraud categories rather than treating them as a subset of ordinary cybercrime. The FBI’s Operation Level Up, focused on cryptocurrency investment fraud, has proactively notified more than 8,000 victims who were actively being scammed and credits the initiative with preventing more than $500 million in additional losses since its 2024 launch, an approach the Bureau has since expanded with Operation Winter SHIELD in 2026 to help organizations harden their defenses against AI-enabled attacks before they succeed.

Perhaps the most sobering figure in the entire federal response effort is the FTC’s own admission about underreporting. While the agency’s Consumer Sentinel Network recorded roughly $16 billion in reported fraud losses for 2025, the FTC’s own December 2025 report to Congress estimated that true losses for 2024 may have reached as high as $81.5 billion, roughly five times the reported figure, because the overwhelming majority of fraud victims, particularly older adults dealing with the embarrassment or confusion that often follows a scam, never file a formal complaint at all.

Frequently Asked Questions About AI Abuse in the US in 2026

How much money have Americans lost to AI-related scams?

The FBI tracked 22,364 AI-related complaints in 2025, tied to $893 million in reported losses, though officials say the real figure is likely higher since AI involvement often goes unrecognized.

What is the total cost of cybercrime in the US in 2026?

The FBI’s most recent full-year data shows $20.877 billion in total reported cybercrime losses for 2025, a 26% increase over 2024.

How common are AI voice cloning scams?

Research shows AI can clone a voice with 85% accuracy from just a three-second audio clip, and roughly 70% of people cannot reliably distinguish a cloned voice from a real one.

How much have older Americans lost to AI scams specifically?

The FBI attributed $352 million in losses among more than 3,100 victims aged 60 and older specifically to AI-enabled scams in 2025.

What is the largest category of AI-related financial fraud?

Investment fraud remains the largest overall cybercrime category at $8.65 billion in 2025, a category increasingly driven by AI-generated fake platforms and AI-cloned endorsement videos.

How much have imposter scams cost Americans?

The FTC reported $3.5 billion in imposter scam losses in 2025, nearly triple the amount reported in 2020.

Is AI making phishing attacks more effective?

Yes. AI-generated phishing emails achieve click-through rates roughly 4 times higher than traditional phishing, and phishing losses nearly tripled, from $70 million to $215.8 million, between 2024 and 2025.

How much fraud actually goes unreported?

The FTC’s own December 2025 report to Congress estimated true 2024 fraud losses may have reached $81.5 billion, roughly five times the amount actually reported to the agency.

What percentage of businesses are affected by AI-related cyber fraud?

The World Economic Forum found 73% of organizations were directly affected by cyber-enabled fraud in 2025, and 87% now cite AI-related vulnerabilities as their fastest-growing cyber risk.

Has the FBI taken action against AI-enabled fraud?

Yes. The FBI’s Operation Level Up has notified more than 8,000 crypto fraud victims and prevented over $500 million in losses since 2024, and the Bureau launched Operation Winter SHIELD in 2026 to further help organizations defend against AI-driven attacks.

Disclaimer: This research report is compiled from publicly available sources. While reasonable efforts have been made to ensure accuracy, no representation or warranty, express or implied, is given as to the completeness or reliability of the information. We accept no liability for any errors, omissions, losses, or damages of any kind arising from the use of this report.

Email ID:

Please enable JavaScript in your browser to complete this form.

Email

Email *

SUBSCRIBE