Skip to content

Use caseFor content agencies

Catch the wrong fact or reference before your client’s reviewer does.

From FinTech whitepapers and DORA or NIS2 explainers to medical affairs decks and pre-MLR drafts, your writers cover technical, regulated subjects on agency deadlines. Lenz identifies the factual statements it can check against primary and authoritative sources and checks that cited studies, statutes, and links exist and support what’s written.

Upload a .docx, get it back with comments and tracked changes.

Example: a regulatory whitepaper“The Digital Operational Resilience Act (DORA) began to apply to EU financial entities on 17 January 2025. EU member states had until 17 October 2024 to transpose the NIS2 Directive into national law. Under NIS2, important entities face administrative fines of up to €10 million or 2% of worldwide annual turnover.”

What Lenz found

FalseNIS2 requires important-entity fine ceilings of at least €7 million or 1.4%.

01Example

Three statements in a regulatory whitepaper. One fine ceiling is wrong.

Deep check · Checked 15 November 2026.

Open a row for the sources it used and the reasoning that weighed them.

Open a deep check to compare the statement with the regulation’s text and sources before editing the draft.

02References

A reference that resolves can still contradict the sentence.

Verify AI citations and check for fake, made-up, or hallucinated references and claims in agency deliverables, whitepapers, and pre-MLR medical content before publishing a draft for regulated clients and agencies.

3 references cited in a pre-MLR draft2 need attention
3/3
DOI not registereddoi.org/10.1038/s41591-024-09912-4 · Checked Crossref, OpenAlex, PubMed“Five-year follow-up results were published in Nature Medicine (https://doi.org/10.1038/s41591-024-09912-4).”
2/3
Does not supportnejm.org · 10.1056/NEJMoa2307563Draft: “Weekly semaglutide 2.4 mg reduced death from cardiovascular causes by 20% compared with placebo.” Source: the 20% reduction applies to the composite primary endpoint (hazard ratio, 0.80); death from cardiovascular causes had a hazard ratio of 0.85 (P = 0.07).
1/3
Supportsnejm.org · 10.1056/NEJMoa2307563 · 17,604 patients aged 45 or older with BMI ≥ 27

Writers and AI assistants frequently cite a real paper or regulation while overstating a secondary endpoint, swapping an entity tier, or pasting a broken DOI.

Lenz resolves DOIs across Crossref, OpenAlex, PubMed, and Semantic Scholar, looks up statutory citations on EUR-Lex, the US Code, the eCFR, UK legislation, and German federal law, and reads linked web pages to check whether the source passage supports the sentence.

Quick checks return in ~15 seconds; deep checks take about a minute or two per statement.

03Deliverables

Built for agencies writing in regulated and technical verticals.

FinTech, cybersecurity & RegTech

Whitepapers, DORA, NIS2, and EU AI Act explainers, benchmark reports, and technical articles. Check statutory thresholds, enforcement fines, breach statistics, and filing figures against primary sources before the client’s CISO or compliance counsel reads the draft.

Healthcare, pharma & MedComms

Medical affairs decks, CME modules, disease-state explainers, and pre-MLR manuscripts. Check trial populations, efficacy endpoints, and study references against PubMed, Crossref, and OpenAlex before your team builds the MLR submission pack.

Executive thought leadership & research

Survey-backed industry reports, ghostwritten bylines, and market briefs. Verify every historical benchmark, competitor figure, and cited study so an account director can hand over the draft with the sources on record.

04In your document

Upload a Word draft, or check inside Google Docs, ChatGPT, and Claude.

nis2-dora-whitepaper.docx1 comment · 1 suggested edit

Under NIS2, important entities face administrative fines of up to €10 million or 2% of worldwide annual turnover at least €7 million or 1.4% of worldwide annual turnover, whichever is higher.

LenzDeep check: False, score 2/10. NIS2 requires important-entity fine ceilings of at least €7 million or 1.4%. Sources and reasoning: lenz.io/c/…

In the Workbench (Word & PDF)

Upload a .docx, a PDF, or an image of a page in the Workbench on desktop. Download an annotated .docx with Word comments on flagged statements and citations (showing the verdict, finding, and top sources) and suggested edits written in as tracked changes you can accept or reject in Word. Review a suggested edit before accepting it: a suggested edit is not itself verified.

Upload a Word draft in the Workbench

In Google Docs, ChatGPT, Claude, n8n, or Zapier

Install the Google Docs add-on to check a draft in a sidebar and apply a suggested correction in place. Or connect Lenz to ChatGPT or Claude with one sign-in, or add an automated check step in n8n or Zapier before a draft moves to client review.

05Scope

What Lenz checks, and what stays with your reviewers.

Checks

The factual statements and cited references in your deliverable, against public sources and registries.

Dates, statistics, regulatory thresholds, clinical trial populations and endpoints, DOIs, PubMed papers, EU acts, US statutes, and public URLs—plus your client’s own reference documents when supplied through the API.

  • Statistics & dates
  • Trial endpoints
  • DOIs & PubMed
  • EU acts & statutes
  • Source support

Doesn’t check

Not a substitute for client MLR sign-off, legal opinion, or brand approval.

Lenz does not approve promotional claims under FDA, EMA, or FCA rules, judge tone of voice or editorial positioning, or guess whether prose was written by an AI model.

  • MLR sign-off
  • Promotional compliance
  • Legal advice
  • Brand voice
  • AI detection

Warranty

Vendors disclaim accuracy. Lenz warrants the result.

A verdict can qualify when a full verification returns True or False with high panel confidence and enough independent sources behind it. Its certificate is a record you can send with the deliverable.

On a paid Pro or Scale plan, every full verification that qualifies carries a contractual warranty, at no extra charge.

How the warranty works

06Pricing

Try free, no card.

Free

$0

100 quick checks or 10 deep checks a month, in any combination.

Try free
Plus

$7.99/ month

500 quick checks or 50 deep checks a month, in any combination.

Or $6.59 a month, billed $79 a year

Choose Plus
Pro · warranty included

$99/ month

5,000 quick checks or 500 deep checks a month, in any combination. The warranty on qualifying verdicts.

Or $83 a month, billed $990 a year

Choose Pro
Scale · warranty included

$399/ month

20,000 quick checks or 2,000 deep checks a month, for production volumes. The warranty on qualifying verdicts.

Or $333 a month, billed $3,990 a year

Choose Scale

07Questions

Yes. Upload a .docx in the Workbench on desktop and download the annotated .docx with Word comments on flagged statements and citations and suggested edits written in as tracked changes. Review any suggested edit before accepting it: a suggested edit is not itself verified.

Yes. Lenz resolves DOIs and academic papers across Crossref, OpenAlex, PubMed, and Semantic Scholar, and statutory citations across EUR-Lex, the US Code, the eCFR, UK legislation, German federal law, and CourtListener, then checks whether the source text supports the sentence. When a page is behind a paywall or login, Lenz marks it as needing a closer look rather than guessing.

No. Lenz catches wrong figures, misquoted endpoints, and broken or unsupported references before submission so your client’s MLR, legal, or compliance reviewers spend their time on judgment, not hunting down misattributed footnotes.

Yes, through the API you can supply the client’s reference documents so Lenz checks the draft against them alongside public sources. Those checks stay private and do not qualify for the warranty.

When no public record settles a statement, Lenz answers “No Public Record” instead of a score. That doesn’t make the statement false.

Every check is private by default. On the Pro and Scale plans you can enable zero retention so submitted text is never written to Lenz’s database.

Not answered here? Talk to us

Send the next deliverable to your client with the facts checked.