Scammers stole nearly $21 billion from Americans through cyber-enabled fraud last year, the FBI reported, as complaints topped one million for the first time. The losses hit older adults hardest, with people 60 and older accounting for roughly $7.7 billion of the total. Separately, the Federal Trade Commission recorded more than $12.5 billion in fraud losses in 2024, driven by investment schemes and imposter cons that are growing harder to detect as criminals adopt artificial intelligence tools to sound and look more convincing.
Why $21 billion in fraud losses demand attention right now
The raw dollar figures are alarming on their own, but the rate at which victims are losing money is accelerating even faster than the complaint count. The FTC found that the share of fraud reports that included a financial loss jumped from 27% to 38% year over year, a shift that suggests scammers are converting a larger proportion of their targets into paying victims. That pattern points to better-crafted pitches, not just more of them.
A key driver is the growing use of AI in fraud delivery. Voice-clone technology and AI-generated profiles now give con artists a way to impersonate trusted contacts or fabricate convincing financial advisors. Federal derivatives regulators have warned about trading schemes that invoke AI “bots” and “algorithms” as marketing hooks, while securities regulators and brokerage watchdogs have highlighted how deepfake audio and video can be used to sell fake investments that appear to come from legitimate firms. Neither enforcement arm has yet published complaint-level data isolating AI-driven schemes from conventional fraud, which means the full scale of the problem is likely larger than current tallies capture.
If AI-enhanced delivery methods continue to improve the success rate of each individual pitch, per-complaint losses in investment and romance scam categories should climb measurably by the next IC3 reporting cycle. The 27%-to-38% loss-rate jump already hints at that trajectory. Criminals who once relied on poorly written emails can now generate polished, personalized messages at scale, reducing the friction that used to protect less attentive targets. Combined with stolen data from previous breaches, AI lets fraudsters tailor messages to a victim’s job, family, or recent purchases, making even cautious consumers more likely to click, respond, or send money.
FBI and FTC data trace $21 billion across investment, imposter, and social media schemes
Investment scams were the single most expensive category, accounting for more than $5.7 billion in reported losses in the FTC’s 2024 data. Imposter scams, where criminals pose as government officials, tech-support agents, or romantic interests, added another $2.95 billion. Those two categories alone represent more than two-thirds of the agency’s total fraud tally, underscoring how persuasive authority figures and financial “experts” can be when they reach people at the right moment.
Social media has become the primary on-ramp for many of these schemes. Nearly 30% of people who reported losing money said the initial contact came through a social platform, generating about $2.1 billion in reported harm. Fake investment pitches, fraudulent online shopping ads, and romance cons all thrive on feeds where users are conditioned to trust recommendations from apparent peers and where paid promotions can make a scam look like a mainstream brand. Once a victim engages, scammers often steer the conversation off-platform to encrypted messaging apps, making it harder for platforms and law enforcement to intervene.
The FBI’s Internet Crime Complaint Center separately tallied nearly $21 billion in cyber-enabled fraud, with complaints surpassing one million. People aged 60 and older bore a disproportionate share, reporting approximately $7.7 billion in losses. Cryptocurrency played a central role in many of those cases. The Justice Department seized more than $6 million in alleged proceeds from a single crypto-confidence scheme in Washington, D.C., illustrating the “pig butchering” model in which victims are groomed over weeks or months before being guided into fake trading platforms that show fabricated profits to keep them depositing more.
Federal enforcement has tried to get ahead of the curve. The FBI launched Operation Level Up in January 2024, a proactive notification program that contacts potential cryptocurrency-fraud victims before they lose more money. Working with exchanges and blockchain analytics firms, agents attempt to flag suspicious transfers in real time and warn consumers that the wallets receiving their funds are linked to known scams. The program has provided periodic updates on the number of victims notified and estimated losses prevented, though granular public metrics on long-term outcomes remain limited and it is too early to know how much behavior change those alerts produce.
Gaps in AI fraud tracking and what consumers should watch for next
The biggest blind spot in the current data is the absence of a dedicated AI category in either the IC3 annual report or the FTC Data Book. Both agencies have flagged AI as a growing threat, but neither breaks out complaint counts or dollar losses specifically tied to AI-generated voices, images, or text. Without that granularity, regulators and researchers cannot measure how much of the loss-rate acceleration is attributable to AI tools versus other factors like cryptocurrency adoption, social media marketing, or pandemic-era shifts in online behavior.
State-level consumer protection offices are beginning to fill in some of the detail with their own alerts about AI voice-cloning used in “grandparent” scams and spoofed video calls in romance or business-email compromise schemes. But those snapshots are not yet standardized or aggregated in a way that can guide national policy. As a result, lawmakers are debating AI-specific disclosure rules and consent requirements with limited evidence about which technologies are actually driving the largest losses.
For consumers, the practical takeaway is to focus less on the technology and more on the tactics. Whether a scammer uses AI or a script, several red flags recur across investment, imposter, and social-media schemes:
- Urgent pressure to act. Demands to send money or share information immediately, especially when paired with threats of arrest, account closure, or missing out on a “guaranteed” return.
- Unusual payment methods. Requests for cryptocurrency transfers, gift cards, payment apps, or wire transfers that are difficult to reverse.
- Requests to move conversations. Efforts to shift from a social platform or email into encrypted messaging or private channels where oversight is lower.
- Too-good-to-be-true promises. Investment offers with high returns and little or no risk, or jobs that pay large sums for minimal work.
- Inconsistent details. Slight name changes, odd email domains, or phone numbers that do not match the organization a caller claims to represent.
Experts advise independently verifying any unexpected outreach that involves money or sensitive data. That means hanging up and calling a known number on the back of a card or from an official website, navigating to a company’s site directly instead of clicking a link, and checking with family members if a supposed relative claims to be in distress. For investment opportunities, consumers can search for disciplinary histories, confirm registration status with regulators, and be wary of pitches that originate on dating apps or social networks rather than through established financial channels.
On the policy front, the lack of AI-specific fraud metrics makes it harder to target interventions, but it does not prevent action. Regulators can require clearer disclosures around automated tools in financial marketing, encourage platforms to label synthetic media, and push banks and crypto platforms to strengthen fraud-detection systems that flag suspicious transfers before funds leave a customer’s control. Over time, adding AI-related fields to complaint forms and standardizing how investigators code those cases would give policymakers a clearer picture of where to invest enforcement resources.
The $21 billion figure is a blunt reminder that online fraud is no longer a niche cybercrime problem but a mainstream economic drain that touches nearly every demographic group. As scammers refine their tactics with AI and other tools, the burden will fall on regulators, platforms, and consumers alike to adapt. Better data, faster alerts, and simple verification habits can’t eliminate the threat, but they can narrow the window in which the next wave of schemes turns from clever pitch into irreversible loss.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
