Skip to main content

Signal briefing / Global Institutional Trends

Google and Stanford researchers launch AI fact-checking tool

Google DeepMind and Stanford University’s SAFE enhance AI chatbots’ responses by fact-checking with 76% accuracy.

Google and Stanford researchers launch AI fact-checking tool

Google and Stanford researchers is a public record based on article evidence, entity context, event links, and relationship context.

A recent development by Google DeepMind and Stanford University introduces the Search-Augmented Factuality Evaluator (SAFE), a tool designed to fact-check long responses from AI chatbots. SAFE employs a multi-step process, including segmentation, correction, and comparison with Google search results, achieving a 76% accuracy rate in verifying controversial facts. This innovation not only enhances accuracy in AI-generated responses but also presents economic advantages, being over 20 times cheaper than manual annotation.

No matter how powerful current AI chatbots are, there tends to exist a much-criticised behaviour providing users with answers that are somewhat convincing but factually inaccurate. Simply put, AI sometimes ‘runs off the rails’ in its responses, even ‘spreading rumours’. Preventing such behaviour in AI large models is no easy task and is a technical challenge. However, according to the foreign media Marktechpost, Google DeepMind and Stanford University seem to have found a workaround.

Also read: OpenAI’s GPT store fails to meet expectations Also read: US federal agencies now required to have chief AI officer The tool is based on the Search-Augmented Factuality Evaluator (SAFE) Researchers have introduced a tool based on large language models the Search-Augmented Factuality Evaluator (SAFE), which can fact-check long responses generated by chatbots. Their research results, along with experimental code and datasets, have now been made public, click here to view.

The system analyses, processes, and evaluates the responses generated by chatbots through four steps to verify accuracy and authenticity: segmenting the answers into individual items for verification, correcting the above content, and then comparing it with Google search results. Subsequently, the system also checks the relevance of each fact to the original question. Researchers created a dataset called LongFact to assess its performance To assess its performance, researchers created a dataset called LongFact containing approximately 16,000 facts and tested the system on 13 large language models from Claude, Gemini, GPT, and PaLM-2.

The results show that in the focused analysis of 100 controversial facts, SAFE’s judgment accuracy reaches 76% upon further review. At the same time, the framework also has economic advantages: it is more than 20 times cheaper than manual annotation.

Member Briefing

Read the full trend briefing

Members read the full briefing with the source notes behind every finding.

Snapshot
Confidence
Confidence score guide
Good confidence (78%)

Published reporting

BackMore Coverage: Global Institutional Trends

Recommended reading

Stanford's 78% is a respondent share, and the base it prints is not its own
Global Institutional TrendsStanford's 78% is a respondent share, and the base it prints is not its own19 Nov 2026
A hundred networks, then the world: Ericsson's unpublished step
Global Institutional TrendsA hundred networks, then the world: Ericsson's unpublished step19 Nov 2026
EU–Canada Digital Trade Talks Are Done. The Agreed Text Is Missing From the Public File.
Global Institutional TrendsEU–Canada Digital Trade Talks Are Done. The Agreed Text Is Missing From the Public File.2 Nov 2026
Thirty Days to Rescue Your Domain. No Shopping for a Rescuer.
Global Institutional TrendsThirty Days to Rescue Your Domain. No Shopping for a Rescuer.1 Nov 2026
The Factory Makes the Phone. The Brand Owns the Decision.
Global Institutional TrendsThe Factory Makes the Phone. The Brand Owns the Decision.1 Nov 2026