Dehype Science headlines, evidence checked
View example Dehype an article
Detect overclaiming Grade evidence Trace sources

Take the hype out of science headlines.

Paste a health or science article and Dehype shows what the underlying study actually supports, what it does not, and whether the claim is clinically actionable.

Source-traced Evidence-graded Claim-audited Not personal medical advice

Free quick checks for basic triage. Paid reports add full source chain, claim audit, caveats, export and citation trail.

Claim gap: High The headline implies patient benefit; the study used animal models.
Amber report

Simple supplement reduces Alzheimer’s damage?

Real preclinical science, but the consumer-health framing outruns the evidence.

Verdict Amber
Evidence grade D
Actionability None
Source confidence High
Claim audit

Article implies: a simple supplement may reduce Alzheimer’s damage.

Evidence supports: arginine reduced amyloid pathology in animal and laboratory models.

Over-framed
Safer summary Promising preclinical work, not proof of human benefit.
Article → PaperSource tracing
Headline → ClaimClaim extraction
Study → GradeEvidence hierarchy
Evidence → ActionClinical relevance
Workflow

From exciting headline to evidence-grade report.

Dehype is built around the real failure mode of science journalism: the paper may be genuine, but the public-facing framing can be too confident.

Paste a science article

Start with the article the reader actually saw: a health headline, press release, supplement claim, or viral science story.

Trace the source chain

Dehype identifies the likely press release, original paper, study design, and whether human outcome data exist.

Audit the claim

The report compares what the article implies with what the underlying evidence can actually support.

Grade the evidence

Get a clear verdict, evidence maturity, missing caveats, actionability judgement, and safer interpretation.

Example output

A report format people might actually pay for.

The value is not a summary. It is the source chain, claim mismatch, missing caveats, actionability and safer wording.

Amber Evidence Grade D Not clinically actionable

Scientists identify simple supplement that greatly reduces Alzheimer’s damage

A real preclinical paper — but the consumer-health framing risks implying benefit that has not been shown in humans.

Evidence summary

Study typeAnimal / mechanistic
Human subjectsNo
Patient outcomeNo
Best readingHypothesis-generating
Safer interpretation

Arginine reduced amyloid-related pathology in animal models. This does not show that arginine supplements prevent or treat Alzheimer’s disease in humans.

Article implies

A simple supplement greatly reduces Alzheimer’s damage.

Evidence supports

Arginine reduced amyloid-related pathology in animal and laboratory models.

Over-framed
Article implies

Potential preventive strategy.

Evidence supports

No human prevention trial or patient-level dementia outcome data were presented.

Premature
Article implies

Already sold as a dietary supplement.

Evidence supports

Availability does not prove effective dose, formulation, duration or clinical benefit.

Needs caveat
View paid-report options
Trust layer

Not a truth oracle. An evidence appraisal workflow.

Dehype avoids simplistic true/false labels. It asks a better question: does the article’s wording match the maturity and limits of the underlying evidence?

Evidence hierarchy Cell, animal, observational, RCT, systematic review and guideline-level evidence.
Claim precision Separates association, mechanism, causality, clinical endpoint and patient benefit.
Overclaim flags Highlights premature supplement, cure, prevention and breakthrough framing.
Source transparency Shows article, press release, paper, citations and source confidence where available.
Use cases

Built for people who need science claims to be precise.

The strongest early audience is not everyone. It is clinicians, writers, educators and serious readers who already care about evidence quality.

Clinicians

Quickly respond to patient-shared articles about supplements, dementia, cancer, diet and drugs.

Journalists & editors

Check whether a headline overstates causality, clinical relevance, effect size or evidence maturity.

Educators

Teach students the difference between real research, press releases and over-interpreted science news.

Careful readers

Find out whether an exciting or worrying article should change your view — or is only hypothesis-generating.

Pricing preview

Start free. Pay when the report needs to be rigorous.

The sensible model is free quick checks, single paid reports, and a professional tier for people who use the workflow repeatedly.

Free

For occasional checks

£0
  • 3 quick checks per month
  • Evidence type
  • Short verdict
  • Basic source match
Start free
Best starting point

Plus

For serious readers

£7.99 /month
  • 30 detailed reports
  • Claim audit
  • Missing caveats
  • Saved report library
Try Plus

Professional

For clinicians, writers and educators

£19 /month
  • 200 reports
  • PDF export
  • Patient-friendly summaries
  • Priority source tracing
Go Pro
Launch concept

Check the next science headline before you believe it.

Dehype is strongest when it stays narrow: health and biomedical headlines, judged through source tracing, evidence maturity and clinical actionability.