Paystub Checker

To detect a fake paystub, reconcile the math first: gross minus deductions must equal net, and YTD totals must add up across the pay periods on file. From there, check the employer is real and consistent with other employment documents, and inspect the stub itself for inconsistent fonts, broken alignment, and edited PDF metadata.

A modern fake paystub detector automates every layer — math reconciliation, template fingerprinting, font and metadata forensics, and AI-generation analysis — and returns a single trust score so HR and underwriting teams can act on the underlying signals.

Gross/net mismatch — gross minus deductions doesn't equal net for the current period.

YTD inconsistencies — year-to-date totals don't reconcile with per-period earnings on file.

Pay period inconsistencies — overlapping or missing periods, irregular pay frequencies, or dates that drift across stubs.

Employer formatting issues — header, address, EIN, or logo styling that doesn't match the employer's real paystubs.

Tax and deduction anomalies — federal/state tax rates, FICA, or benefit deductions that don't match the jurisdiction or earnings.

Font and layout anomalies — mixed fonts, drifting kerning, broken column alignment, or off-template spacing.

Metadata edits — modified PDF Producer, ModDate, or incremental updates that betray after-the-fact edits.

Suspicious formatting — round-number earnings, unrealistic deductions, or layouts copied from public stub generators.

Inflated gross or net earnings vs the role, employer, and region. Fictitious employer names and addresses that don't match employer registries or other employment documents.

Math that doesn't reconcile — gross minus deductions doesn't equal net, or YTD doesn't add up across the periods on file.

Reused templates from public stub generators, identifiable through font and layout fingerprints.

Fully synthetic AI-generated paystubs that match a known template but reference no real employer or pay history.

Edited PDF metadata, missing XMP entries, or incremental update chains that betray after-the-fact edits.

AI-generated paystub detection is the highest-leverage signal in modern HR and rental screening workflows. Generative tools now produce paystubs that look convincing at a glance — TrueDoc evaluates AI-generation signatures, layout regularities, and rendering artifacts common to current text and image models.

For HR teams running employment document fraud checks at scale, the same fake paystub detector pipeline is available through the dashboard or via the document fraud detection API — see HR & Employment Document Verification for a workflow-level overview.

Public paystub generators and AI tools produce stubs that look convincing at a glance — matching headers, line-item structure, and YTD math. TrueDoc's fake paystub detection works regardless of template, including ADP, Paychex, Gusto, QuickBooks, and fully custom layouts.

Every stub is scored for both forensic signals (font, alignment, metadata, AI generation) and math integrity (gross, deductions, net, YTD).

Generative-AI tooling has lowered the bar for paystub forgery. TrueDoc evaluates AI-generation signatures, layout regularities, and rendering artifacts common to current-generation text and image models.

When combined with broader employment document fraud signals — fictitious employers, recycled stubs, inconsistent identifiers — TrueDoc gives recruiters, HR teams, and underwriters a single Document Trust Score per submission.

Paystub fraud is one of the most common forms of income verification fraud — feeding into rental approvals, consumer lending, BNPL, and remote hiring. TrueDoc plugs into existing HR document verification workflows via the dashboard or the document fraud detection API.

Flagged stubs are returned with structured evidence so reviewers can act on the underlying signals rather than re-inspecting the document from scratch.

TrueDoc analyzes documents from any bank, employer, or country. Below is a format guide for the three most common US payroll providers — used here as concrete examples of US-specific red flags surfaced during analysis.

ADP stubs use a boxed header with the employer name, address, and pay period, then two side-by-side blocks for 'Earnings' (rate, hours, current, YTD) and 'Deductions / Taxes' (current, YTD), followed by a 'Net Pay' summary and a check-stub tear-off. Gusto stubs use a cleaner single-column layout with a green Gusto footer, an Earnings table above a Deductions and Taxes table, and explicit 'Employer Contributions' and 'Employee Taxes' sections. Paychex stubs use a dense multi-column table with 'Hours & Earnings' on the left and 'Taxes / Deductions' on the right, plus a 'YTD' column on every line and a footer with the Paychex service address. Real stubs from all three keep column alignment identical page-to-page and use provider-specific fonts (ADP typically Arial-family, Gusto typically Inter/geometric sans, Paychex typically Helvetica-family).

US-specific math checks TrueDoc runs on every stub: gross minus total deductions must equal net for the current period; YTD gross, YTD taxes, and YTD net must reconcile with the sum of periods on file; Social Security withholding must equal exactly 6.2% of Social Security wages up to the annual wage base ($168,600 for 2025, $176,100 for 2026); Medicare must equal exactly 1.45% of Medicare wages, plus an additional 0.9% on wages above $200,000 single-filer; federal income tax withholding must be plausible against gross and filing status given IRS Publication 15-T brackets; and state withholding must be plausible for the claimed state (zero for TX/FL/WA/NV/TN/SD/WY/AK/NH, flat-rate for PA/IN/CO/UT/MI/IL/NC, bracketed for CA/NY/NJ/OR/HI, etc.).

US paystub red flags: FICA rates that aren't exactly 6.2% and 1.45%, Social Security wages exceeding the annual wage base with continuing 6.2% deductions, state tax withholding on a paystub claiming a no-income-tax state, YTD figures that decrease period-over-period, round-number gross amounts every period, EIN format that doesn't match the XX-XXXXXXX pattern, an employer address in a state that doesn't match the state tax jurisdiction, and provider-branded footers (ADP / Gusto / Paychex) paired with the wrong fonts or column layout — a classic tell of a generic stub-generator template mislabeled as an ADP or Gusto stub.