OpenAI and Anthropic have set up dedicated operations to guard against attempts to use artificial intelligence to disrupt the Midterm elections. Google and Meta have also pledged efforts to prevent their tools from being used to undermine confidence in the democratic process.
At an OpenAI office in San Francisco, more than a dozen people are working around the clock to monitor potential election-related misuse of ChatGPT. The operation is a major expansion of the company’s work ahead of the 2024 presidential election, when AI tools were less capable and ChatGPT had one-fifth of its current one billion users.
Josh Lawson, OpenAI’s head of civic and strategic risk, said the range of threats has grown beyond text. Potential risks now include deepfakes, cyberattacks and foreign or domestic influence operations that can be created and launched with prompts.
Anthropic, the company behind Claude, has established its first election “war room,” with analysts working in San Francisco, New York and Washington. The companies’ measures come as AI becomes more embedded in online social and political life, amid criticism that technology firms have not done enough to limit malicious content.
Sources: The New York Times Business.