Skip to content
Submit incident
AII-1001-59373
ReportedReported by fewer than three independent publishers, with no official finding.Supporting sources are pending verification.

Google AI Overviews spread inaccurate health information

The Guardian investigation found that Google AI Overviews provided inaccurate health information at the top of search results, including wrong advice on pancreatic cancer dietary recommendations and incorrect liver function test information. Health experts warned these errors could potentially harm patients, particularly those seeking medical information during health crises.

2 January 2026 Β· Google Β· Reliability and hallucination
2 January 2026
AII-1001-59373
Quick facts
Organization
Google (subject)
Category
Reliability and hallucination
Country
United Kingdom
Date
2 January 2026
Drafted by
Raidu Incident Detection

What happened

The Guardian investigation found that Google AI Overviews, which appear at the top of search results, served inaccurate health information in multiple cases.

In one case, Google wrongly advised people with pancreatic cancer to avoid high-fat foods, the opposite of medical recommendation. In another case, the system provided incorrect information about liver function tests. Google searches also returned completely wrong information about women's cancer tests.

Reported impact

Experts said the pancreatic cancer advice was 'really dangerous' and may increase the risk of patients dying from the disease. Experts described the liver function test information as 'alarming' and said it could leave people with serious liver disease wrongly thinking they are healthy.

Directors of health charities said inaccurate information could seriously harm people's health, particularly those in moments of worry and crisis who might dismiss genuine symptoms.

Index analysis

Google stated that the AI Overviews are 'helpful' and 'reliable' and that many examples shared were 'incomplete screenshots' that linked to well-known, reputable sources.

Classification

Organization
Google (subject)
Category
Reliability and hallucination
Industry
Healthcare, drugs and biotechnology
Country
United Kingdom
Harm status
Potential harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Physical injury, Public interest
Harmed
Consumers, General public
Failure modes
Hallucination, Unsafe recommendation
AI systems
AI Overviews

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident