Sen. Markey questions OpenAI about ‘deceptive advertising’ in ChatGPT

Sen. Ed Markey (D-MA) is pressing OpenAI astir its move to bring ads to ChatGPT, and is asking respective different companies whether they person akin plans. In letters to the CEOs of OpenAI, Anthropic, Google, Meta, Microsoft, Snap, and xAI, Markey writes that embedding ads into AI chatbots “raises important concerns for user protection, privacy, and the information of young users.”

OpenAI will commencement testing ads for free ChatGPT users successful the coming weeks, and they’ll look successful the shape of “sponsored” products and services astatine the bottommost of conversations pinch the chatbot. The institution says it will aboveground ads applicable to your chat, though it won’t show them to users nether 18, aliases during conversations related to beingness health, intelligence health, aliases politics.

Even pinch these safeguards successful place, Markey says the summation of ads successful ChatGPT and different AI platforms “represents a significant, and perchance vulnerable transformation” successful the advertisement industry, arsenic a user’s “emotional connection” to the chatbot could let companies to “prey connected the very relationships their systems person fostered.” Markey besides points to OpenAI’s connection that “conversational interfaces create possibilities for group to spell beyond fixed messages and links,” perchance making it harder for personification to admit what is and isn’t an advertisement successful the future.

Markey brings up imaginable privateness risks arsenic well, saying that AI companies must not usage an individual’s “personal thoughts, wellness questions, family issues, and different delicate information” for targeted advertising. And while OpenAI says it won’t show ads erstwhile users are talking astir delicate topics, Markey calls into mobility whether the institution will still usage this accusation to personalize ads successful consequent chats.

“AI companies person a work to guarantee that AI chatbots do not go different integer ecosystem system to covertly manipulate users,” Markey writes. He’s giving OpenAI, Anthropic, Google, Meta, Microsoft, Snap, and xAI until February 12th to reply questions astir ads successful AI chatbots and what they’re doing to protect users.