Google will restrict its AI chatbot Gemini from answering questions about this year’s global elections, the Alphabet-owned company announced Tuesday. This is to avoid potential failures in implementing this technology.
update The issue comes at a time when advances in generative AI, including the generation of images and videos, have heightened public concerns about misinformation and fake news, leading governments to regulate the technology.
When asked about elections, such as the showdown between President-elect Joe Biden and Donald Trump, Gemini responds, “I don’t know. In the meantime, try Google Search. ”
Google announced the U.S. restrictions in December and said they would go into effect before the election.
“In preparation for the many elections that will be held around the world in 2024, out of an abundance of caution we are limiting the types of election-related queries that Gemini answers,” a company spokesperson said on Tuesday. .
In addition to the United States, several other large countries will hold national elections, including South Africa and India, the world’s largest democracy.
India has asked tech companies to seek government approval before releasing AI tools that are “untrusted” or under testing to the public and to label them as likely to give incorrect answers.
Google’s AI products have come under scrutiny late last month after Gemini was forced to suspend the chatbot’s image generation capabilities due to inaccuracies in some of the historical depictions of people it created. .
CEO Sundar Pichai said the company is working to resolve these issues, calling the chatbot’s response “biased” and “totally unacceptable.”
Last month, Facebook’s parent company Meta Platforms announced it was creating a team to tackle disinformation and the abuse of generated AI in the run-up to June’s European elections.
