Mark Cuban said the Trump administration needs to crack down on ads in AI models

"Shark Tank" star Mark Cuban said on Saturday that the White House should "make it illegal for AI models to offer advertising."

Cuban said in an X post addressed to David Sacks, the White House's AI and crypto czar, that the administration should "examine referral fees as well."

"The last thing we need is to have algorithms designed to maximize revenue driving LLM output and interactions," Cuban wrote.

"They are already recommending brands and we don't know if they are getting paid for it. We need to have learned our lessons from algos in social media," he added.

Cuban said in a subsequent post on Saturday that he would be willing to accept advertising on AI models if they are "identified as an ad" and kept "completely independent from the user generated chats."

Cuban's proposal comes just days after the Trump administration unveiled its 28-page "AI Action Plan" on Wednesday. Back in January, President Donald Trump had signed an executive order calling for "existing AI policies and directives that act as barriers to American AI innovation" to be revoked.

Trump has adopted a relatively light-touch approach toward AI regulation compared to his predecessor, President Joe Biden. In October 2023, Biden signed an executive order demanding greater transparency from companies developing AI tools.

Trump's new "AI Action Plan" proposed withholding federal funding from states that want to impose "burdensome" AI regulations.

Cuban and the White House did not respond to requests for comment from Business Insider.

Social media déjà vu

Cuban's worries may not be unfounded. Major AI players have been deepening their leadership bench with former executives from social media platforms like Facebook and Instagram.

In May, OpenAI chief Sam Altman said he had hired Fidji Simo, the CEO and chair of Instacart, to serve as OpenAI's new CEO of applications. Before she joined Instacart, Simo worked at Meta, where she oversaw Facebook's app and advertising products.

Last year, OpenAI hired Kevin Weil as its chief product officer. Weil was previously vice president of product at Instagram and senior vice president of product at Twitter.

OpenAI's rival, Anthropic, made a similar move in May 2024 when it hired Mike Krieger, cofounder and former CTO of Instagram, as its chief product officer.

Cuban has long warned about the risks and dangers that could come with AI tools like chatbots. He told comedian Jon Stewart in a podcast interview that aired in 2023 that online misinformation "is only going to get worse" with the proliferation of AI tools.

"Once these things start taking on a life of their own, it will be difficult for us to define why and how the machine makes the decisions it makes, and who controls the machine," Cuban said.

Last week, Cuban wrote in an X post that he expects AI companies to hoard talent and intellectual property to stay ahead of their competitors.

"If you create valuable IP, encrypt and silo it. Let companies bid on it. Or just use it for your own behind a paywall model. IP is KING in an AI world," Cuban wrote on July 20.