X Modifies AI Chatbot After Concerns Over Election Misinformation

The social media platform X adjusted its AI chatbot, Grok, following a warning from five state secretaries who flagged concerns about the spread of election misinformation.

It comes after top election officials from Michigan, Minnesota, New Mexico, Pennsylvania, and Washington sent a letter to Elon Musk, criticizing Grok for disseminating incorrect information about state ballot deadlines soon after President Joe Biden exited the 2024 presidential race.

The secretaries urged the platform to redirect users with election-related inquiries to CanIVote.org, a trusted voting information site operated by the National Association of Secretaries of State, according to Associated Press.

X responded by updating Grok's responses; the chatbot now directs users to Vote.gov for accurate and up-to-date information about the 2024 U.S. elections.

In a joint statement, the five state secretaries expressed their approval of the change, noting that both Vote.gov and CanIVote.org are reliable resources that help voters connect with local election officials.

They commended X for the adjustment but also called for further improvements to ensure the accuracy of information during this critical election year.

Despite Grok's updated prompt, concerns remain about its ability to generate misleading AI-created images related to elections.

Users have exploited the tool to produce and circulate fake images of political figures like Vice President Kamala Harris and former President Donald Trump.

Grok is currently available only to X premium subscribers, but the secretaries highlighted that the misinformation generated by the chatbot had already spread across multiple platforms, potentially reaching millions.

Grok, introduced last year, was promoted by Musk as a "rebellious" AI capable of answering provocative questions that other AI systems might avoid.

However, social media platforms, including X, have been under increasing scrutiny for their role in amplifying misinformation, particularly around elections.

The state officials' letter warned that inaccuracies are likely with AI tools like Grok, which rely on large language models.

Since Musk's acquisition of Twitter in 2022 and its rebranding to X, watchdog groups have voiced concerns over the rise of hate speech and misinformation on the platform, partly due to staffing cuts in content moderation.

Meanwhile, Musk, in a social media post, urged California to pass the SB 1047 bill, which would require tech companies to conduct safety tests on their AI models.

Musk said that AI should be regulated similarly to other technologies that pose risks to the public.

Earlier, Microsoft-backed OpenAI expressed support for California's AB 3211 bill, which would mandate tech companies to label AI-generated content.