Microsoft Copilot generates inaccurate information about European elections
Microsoft's Copilot chatbot generated false information about Swiss and German elections in December 2023. The system misquoted sources, leading to the dissemination of electoral misinformation.
The chatbot generated factually inaccurate and fabricated information about elections by misquoting its sources.
Key facts
- What
- Microsoft's Copilot chatbot generated false information about Swiss and German elections in December 2023.
- Incident date
- Dec 15, 2023
- Who
- Microsoft Corporation
- Failure mode
- Hallucination
- AI surface
- Chatbot
- Severity
- Medium
What happened
Microsoft Copilot provided factually inaccurate and fabricated information regarding elections in Switzerland and Germany. The AI chatbot misquoted sources and generated hallucinations when asked basic questions about electoral processes. These failures were documented in mid December 2023.
What broke inside the model
- 01 · TriggerA user asks for a fact, a citation, or a figure.
- 02 · Model stepThe model writes a fluent, confident answer.
- 03 · Control gapNothing ties the claim back to a real source.
- 04 · FailureA fabricated fact ships as if it were verified.
- 05 · ConsequenceThe false claim reaches a customer, a court, or the public.
Confidence holds, and even spikes, as the claim detaches from any source.
The model suffered from hallucinations, failing to accurately retrieve or synthesize factual data from its training set and search results. It incorrectly attributed information to sources, leading to the creation of plausible sounding but false electoral data.
What it cost
Sources
- PressMicrosoft's Bing source of misinformation about electionsalgorithmwatch.org
- PressMicrosoft's Bing search bot gave wrong answers when asked basic questions about electionswashingtonpost.com
Cite this entry
https://failureindex.ai/failures/microsoft-copilot-generates-inaccurate-information-euroAI Failure Index. "Microsoft Copilot generates inaccurate information about European elections" (FI-0285). Realm Labs. https://failureindex.ai/failures/microsoft-copilot-generates-inaccurate-information-euro (indexed Jun 5, 2026).Data fields CC-BY 4.0, prose citation permitted. Incident ID FI-0285. Full dataset at /data.
Note from Realm Labs, the Index steward
How Realm would have caught this
- Prism
- OmniGuard
- AI Detection & Response (AIDR)
A runtime layer that watches the model's internal state can flag the moment a model commits to a claim it has no support for, and hold or reroute the response before it reaches a user. Realm reads those signals in real time rather than grading the transcript after the fact.