The picture, which featured a sunny infinity pool in Bali with a brand-tagged swimsuit caption, went live on a Thursday. By the weekend, it had received several hundred thousand likes. The visual was satisfying to the followers who were scrolling past it. It was a more specific data point in an ongoing assessment of whether the income that funded the vacation appeared anywhere on a 1040, according to an IRS computer trained to link publicly apparent lifestyle signs versus filed tax returns.
This isn’t speculative. In the creative economy, this is becoming how federal tax enforcement operates, and the majority of influencers who share their sponsored content and opulent travels have little to no awareness that their audience is part of a system that is secretly keeping score.
| Agency | Internal Revenue Service (IRS) — deploying machine learning and AI tools, largely funded by the $80 billion Inflation Reduction Act, to identify tax non-compliance among high-net-worth individuals and content creators |
|---|---|
| Funding Source | Inflation Reduction Act — allocated $80 billion to IRS enforcement and modernisation, a significant portion directed toward AI-driven audit infrastructure targeting high earners |
| Social Media Monitoring | AI algorithms scan publicly available posts on Instagram, TikTok, and YouTube — comparing lifestyle evidence (luxury purchases, travel, brand partnerships) against reported income figures |
| Crypto Non-Compliance Rate | IRS estimates a 75% non-compliance rate among taxpayers using digital currency exchanges — AI is being used to trace underreported crypto income, a common revenue stream for online creators |
| Income Threshold Flagged | AI models flag individuals earning over $1 million with more than $250,000 in recognised tax debt — these cases are escalated for comprehensive human-reviewed audits |
| Third-Party Data Matching | Automated systems match reported income against 1099-K forms, bank data, and financial institution records — discrepancies exceeding $1 million are flagged without manual initiation |
| Large Partnership Focus | IRS developed a dedicated AI model for large partnership tax documents — sending hundreds of compliance alerts to partnerships with significant balance sheet discrepancies identified algorithmically |
| Legal Note | IRS data use is governed by federal privacy law — monitoring is limited to publicly available information and third-party financial data already required to be reported to the agency |
The Inflation Reduction Act, which allocated $80 billion for the agency’s enforcement and modernization operations, greatly hastened the IRS’s long-term shift toward machine learning. A significant portion of that money has been invested on AI infrastructure, which includes systems that can find statistical anomalies among the millions of returns the agency processes each year, scan enormous datasets, and compare public records with financial forms.
High-net-worth individuals, major partnerships, and—increasingly—content creators whose income comes from brand deals, sponsorships, digital products, and cryptocurrency—often through unofficial channels that don’t always produce the paper trail that a traditional salary does—are the targets that are getting the most attention.
In theory, the lifestyle-versus-income approach is fairly simple. When an influencer posts about a new Range Rover, a renovation on a second property, and a number of first-class foreign flights while reporting a yearly salary of $180,000, they generate a set of data points that collectively point to a disparity. The AI only needs to identify the pattern as statistically significant enough to demand further investigation; it does not need to demonstrate the gap like a human auditor would.
The file then advances in the queue. When a human reviewer finally examines it, the algorithm has already completed the task of locating the discrepancy and comparing it to thousands of profiles that are similar. The technology selects only the highest-risk cases for the complete audit process, reviewing considerably more returns than the IRS could ever manage by hand.
Since the IRS has been most forthright about the scope of the issue, the cryptocurrency angle merits special attention. In a creator economy where cryptocurrency payments for sponsored content, NFT sales, and platform tips have become commonplace, the agency’s estimate of the non-compliance rate among taxpayers using digital currency exchanges is about 75%. This figure would be astounding in any industry.
By comparing blockchain activity with known wallet addresses and exchange-reported data, AI technologies are now being used to track these transactions back to individual filers. Large partnerships with balance sheet problems found by algorithmic review have also received compliance notifications from the IRS, which are signals rather than official audit notices. It is a more compassionate variation of the same procedure, intended to promote voluntary rectification prior to the involvement of the formal machinery.
Observing this unfold, it seems as though the creative economy has been functioning for a number of years under a set of enforcement presumptions that are no longer entirely valid. For a while, the informal nature of influencer income—the brand contract paid through an LLC, the company-sponsored trip booked directly, the cryptocurrency transmitted to a wallet—felt like a structural buffer against the kind of scrutiny that ordinary employees encounter when they are paid. That buffer has significantly shrunk.
The IRS has made it clear that creators must manage their platforms like businesses, keeping track of all payments, sponsorships, and stated expenses with the same diligence that a small factory would devote to its accounting. How many people have taken in that message is yet unknown. For its part, the algorithm isn’t waiting to learn.