OpenAI blocks ChatGPT-powered political bot, citing misuse in upcoming elections

OpenAI took action by banning the developer of a bot, Dean. Bot, which was mimicking Democratic presidential hopeful Rep. Dean Phillips. The bot was created by Silicon Valley entrepreneurs Matt Krisiloff and Jed Somers, who supported Phillips through a super PAC called We Deserve Better. The bot utilized OpenAI’s ChatGPT conversational software.

OpenAI suspended the account of the AI startup Delphi, which had been contracted by the super PAC to build the bot, in response to the perceived misuse of its AI tools in a political campaign. OpenAI stated that the suspension was due to violating its API usage policies, prohibiting the use of its technology in political campaigning, or impersonating individuals without consent.

The bot, powered by ChatGPT, could interact with voters in real time through a website. Despite including a disclaimer stating that it was an AI tool and not the real Dean Phillips, researchers raised concerns about the potential harm such technologies could cause elections. There were worries that, even with disclaimers, voters might be lulled into accepting a dangerous tool.

In response to inquiries from The Washington Post about OpenAI’s prohibitions, the developers stated that they had asked Delphi to remove ChatGPT from the bot and instead use open-source technologies. However, the bot remained publicly available without ChatGPT until Delphi took it down in response to the account suspension.

Matt Krisiloff, a former chief of staff to OpenAI CEO Sam Altman, had no further comment, and Delphi did not immediately respond to a request for comment. The suspension and actions taken by OpenAI highlight the ongoing challenges and ethical considerations surrounding using AI tools in political campaigns.

More here.