OpenAI will keep a close eye on the midterm elections after launching what is being called a war room.
It is officially called an Election Operations Center and will be active until all votes are counted, Politico reported.
The initiative will help identify the spread of misinformation and other ways the election could be disrupted.
“We’ve put considerable effort [into] ensuring we help users get reliable voting information, combat misuse of our tools and support election defenders,” Josh Lawson, OpenAI’s head of civic and strategic risk, told Politico. “The Elections Operations Center builds on that yearslong work, bringing our teams together to respond quickly to emerging issues and enforce our policies as people head to the polls.”
Lawson told NBC News that it will “helps support users seeking correct information, reliable information, about the when, where, how to vote process, but also to combats misuse.”
He said OpenAI built safeguards that will keep “bad actors” from using the system to get their content out to more people more quickly. That includes deepfakes and cartoonification, or “photorealistic voter interference into cartoons so it can’t be used for deceptive purposes,” Lawson explained. Politico said the cartoonification would be used instead of the system making a photorealistic image when prompted to create an AI photo of something such as long voting lines.
He said OpenAI has launched partnerships with The Associated Press and Democracy Works to source “helpful information, reliable information.”
Lawson said the company is monitoring for misuse surrounding Election Day.
As for results, ChatGPT will be using the AP for calling elections.
NBC News reported that Anthropic will also be battling misinformation surrounding Election Day.
But the company noted that most Claude.ai users don’t use the platform for politics. It found that of all the chats, only 5.75% touch on politics or social issues, with 0.51% asking about elections or voting.
Asking who to vote for accounts for only 0.01% of all Claude conversations.
Still, the company also rolled out safeguards before today’s election.
“Ahead of Election Day, we’ve added two ways for people to get reliable information directly in their conversations with Claude. As we shared in April, when people ask Claude.ai about voter registration, polling places, or election dates, we’ll continue to display a banner linking to TurboVote, a nonpartisan resource from Democracy Works,” Anthropic said.
It also said the partnership with Democracy Works would let Claude use that organization’s information in its responses.
Politico noted that other tech companies, such as Meta, have in the past set up election centers.
©2026 Cox Media Group
