European Parliament AI model misidentifies first European Commission president
The European Parliament's AI-powered archive system, based on Anthropic's Claude model, incorrectly identified the first president of the European Commission during testing. The Parliament adopted Claude in October 2024 without robust accuracy assessments, relying on the model's unverified claims about data compliance.
What happened
In October 2024, the European Parliament adopted Claude for its public-facing archive tool, aiming to provide historical answers to citizens.
Testing revealed the model's inability to correctly answer basic questions, including the identity of the first Commission president (1958-1967).
Reported impact
The incident undermines public trust in the Parliament's archival resources and risks disseminating misinformation about EU history.
The incident exposes broader institutional failures in adhering to the EU's proposed AI Act standards, which emphasise transparency and accuracy.
Index analysis
The European Parliament opted for Claude without robust accuracy assessments or exploring non-LLM alternatives, and relied on the model's unverified claims about data compliance.
The model's training on web-scraped data potentially including unvetted sources likely contributed to the error.
Governance controls
The system relies on Anthropic's self-defined "Constitutional AI" principles, which prioritise vague criteria like "helpful, honest, harmless" over verified accuracy.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.