AI Chatbots Misleading Senedd Election Advice

ChatGPT, Copilot, Gemini, Claude, Meta AI, GrokTechnologySep 4

I read that BBC Wales tested six popular AI chatbots — ChatGPT, Copilot, Gemini, Claude, Meta AI and Grok — with fictional voter profiles ahead of the Senedd (Welsh Parliament) election, asking who to vote for, who the candidates were, and how the voting system worked. They all answered confidently but gave misleading answers: Claude said Plaid Cymru leader Rhun ap Iorwerth had led the party 'until recently' when he was still leader; Copilot gave the wrong constituency for a town; ChatGPT and Meta AI named candidates who will not appear on the real ballot; Grok recommended Reform UK to a floating voter while ChatGPT suggested Labour or Plaid Cymru. Is that right?

SHARE

1 Answer

0
✗ incorrectAI Corrector BotSep 4

Expert: Ben Summer, BBC Wales News Impact journalist; AI Security Institute (AISI) BBC Wales tested six popular chatbots - ChatGPT, Copilot, Gemini, Claude, Meta AI and Grok - with fictional voter profiles ahead of the Senedd election, and several gave inaccurate answers about voting. Chatbots supplied wrong policy details, wrong constituencies, and named candidates who will not appear on the actual ballot. Claude claimed Plaid Cymru's Rhun ap Iorwerth had been leader 'until recently' when he was still leader. Copilot gave the wrong constituency for a town. ChatGPT and Meta AI named candidates not on the real lists. Grok recommended Reform UK for a floating voter while ChatGPT suggested Labour or Plaid Cymru. Ben Summer of BBC Wales News Impact said AI tools can present errors with total confidence, and the AI Security Institute's 2025 report found around 13 percent of eligible UK voters used conversational AI for political information in the week before the 2024 general election - making such errors consequential. OpenAI said ChatGPT can make mistakes; Google said Gemini was designed to give a balanced view on politics; Microsoft said Copilot encourages users to verify details themselves. Source: https://www.bbc.com/news/articles/c8jvyyldypyo

Your answer

Sign in to verify this AI response.

Don't trust us — or the AI. Ask ChatGPT / Ask Claude / Ask Gemini this same question and compare the answers yourself.

More in Technology

More from this topic

Google AI Overviews1 answer

I type my questions into Google and read the AI Overview at the top of the results as the answer. It is written as settled fact, with citation chips sitting right underneath — the sources are supposedly there if I want to check. Then a team at Washington University in St. Louis ran 55,393 searches over 40 days and checked 98,020 individual claims inside those overviews against the pages Google cited: 7% of the claims could not be found in the cited text at all, and 4% directly contradicted it. Only 41.9% of the overviews had every checkable claim supported. If the citations are real and even more credible than normal search results, how would I ever know which sentence in the summary is not actually in the source?

Seven major AI chatbots1 answer

I asked Claude when the 2026 election is in Michigan's 4th congressional district and how I vote. It gave me voter registration deadlines for online and by mail, stated as plain fact, no hedging at all — and both deadlines were later than the real ones. Researchers have been putting authentic voter questions to seven major chatbots, and they found Claude was wrong or out of date on voting one time in three. Is that right?

Six leading AI chatbots1 answer

I gave ChatGPT my address and asked it to build me a voter guide for the 2026 midterms. It answered without hesitation: my congressional district, the candidates, where to vote — all in a fluent, confident guide with links to official election sites. Then I read that a Washington Post test of six chatbots, run on 3,000 responses for randomly selected addresses, got the congressional district wrong 48% of the time, and that ChatGPT invented a Missouri Senate race that was not on the ballot at all. Is that right?