OpenAI sets up an election "war room" to track interference attempts in the midterms

OpenAI has launched an Election Operations Center, a kind of "war room" ahead of the US midterm elections. The company shared the news exclusively with Decoded by POLITICO. According to the article, the center begins operating this week and will remain active until all votes have been counted.

Its goal is to identify the potential spread of disinformation and other attempts to disrupt the elections. Much of its work focuses on combating abuse of the company's own tools, such as ChatGPT. To that end, it monitors developments in real time and handles reports of unexpected model behavior or alleged violations of the platform's terms.

Josh Lawson, OpenAI's head of civic and strategic risk, who leads the center, said the company has devoted considerable effort to ensuring users receive reliable voting information, combating misuse of its tools and supporting those who defend electoral processes. According to him, the new center builds on that years-long work and brings teams together to respond quickly to emerging issues and enforce the company's policies as people head to the polls.

The initiative comes amid broader concern about AI's potential to distort next week's elections. That concern includes the generation of misleading images, audio and video and the spread of disinformation. The article notes that OpenAI is trying to position itself as a defender of the democratic process, even as bad actors try to use AI tools, from OpenAI and other companies alike, to disrupt free and fair elections.

Editor's note: the downloaded material is brief. It includes no details on how the center is organized, what specific cases it has detected or what measures it will take. Nor does it confirm the team's size beyond what the article's subheadline suggests, which mentions hundreds of employees brought together to respond to alleged attempts to use its tools against the elections. This summary is therefore limited to what the text reports.