Senate Hearing Tackles Harms Caused by Social Media Algorithms

The day after senators grilled Instagram CEO Adam Mosseri on the negative effects of his platform on teen girls during a Senate Commerce subcommittee hearing, another subcommittee heard from witnesses about online platforms’ use of technology to manipulate their users’ experiences.

From their questioning, members of the Senate Commerce Communications, Media, and Broadband Subcommittee are looking for ways to make platforms accountable when the companies put optimizing user engagement – which generates ad revenue and user data – over the harmful effects to those users.

“Online platforms collect troves of private demographic and behavioral data about us all and then use it to target us with ads, recommendations, and other content based on our perceived interests and vulnerabilities. That data collection is pervasive, and may be invasive and extractive,” Jessica J. González, Co-Chief Executive Officer of Free Press Action, said in her written testimony. “[T]he companies collecting that data have little to no accountability to the American people.”

Rose Jackson, Director of the Democracy & Tech Initiative, Digital Forensic Research Lab, Atlantic Council, warned there are foreign policy and national security implications arising from the platforms’ behavior.

“Authoritarian countries like China and Russia are taking an entirely different approach. They are proactively driving an alternative vision for the internet that is incompatible with our constitutional democracy and universal rights,” Jackson said. “[They benefit] from a fragmented approach to the internet. If they can drive a wedge between the world’s democracies, they can reinforce their control at home while limiting civic space online for everyone.”

Dean Eckles, associate professor of marketing at the MIT Sloan School of Management, testified that there are no easy answers about what the algorithms should do. “Platforms often struggle [to] quantify the decisions they must make,” he said. “We lack clear evidence about harm, [and] simple chronological ranking can make” some things worse.

James Poulos, executive editor of the American Mind from the conservative Claremont Institute, offered an alternative – but still negative – view. “The meme that algorithmic harm comes from greedy companies” limits understanding of what social media is about, he said. “It’s about control,” tolerating only some opinions. He blamed an emerging “cyborg theocracy” that believes “we’ll merge with our computers and become like gods.”

“Giving consumers control over their own data is an overwhelming priority,” Sen. Ben Ray Luján (D-N.M.), chairman of the subcommittee, said. “Eighty-six percent of Democrats and 81% of Republicans want privacy to be a priority for Congress.”

Jackson said people are aware that something they query or click on one platform then shows up on another platform, but they don’t understand how all the social media platforms cross-pollinate. Congress needs to focus not on content but on “the mechanisms of the platform itself and how it spreads from platform to platform,” she said.

Eckles said that while the social media platforms have massive amounts of data on their users’ behaviors and preferences, they can’t yet predict users’ actions, though they work hard to influence them. “It may be hard for consumers to understand how purchasing information could be combined with data about [other people’s behavior] to shape” what they’re shown online.

“It’s inherent in the nature of this technology … to influence people,” Poulos said. “The printing press, radio, TV … This is how it works – whoever controls the communications infrastructure” controls the public. “What’s really going on here is amassing information in databases so that large groups of people can be conditioned to behave in certain ways.”

Sen. Amy Klobuchar, (D-M.N.), noted that in the previous day’s hearing with Instagram, the CEO had said the major increase in his marketing budget was aimed at wooing teens to join the platform. “Algorithms touch every aspect of our lives, we know that, and it’s typically not very pretty,” she said.

“[We’re] getting after exclusivity, self-preferencing, predatory behavior,” Klobuchar said. “The laws haven’t caught up. My thought has always been that competition is good for markets. [How can] this innovation … actually cause monopolies and stunt competition.”

“I started a small tech company to help citizens become more engaged. I didn’t have much choice but to use Facebook and Google to advertise,” Jackson said. “I think there are serious consequences for competition [in the U.S.], but everyone feeds into this system … All these issues are interconnected because this system feeds itself.”

Eckles said that Facebook’s attempt to rebrand itself as Meta is intended to ease the bad publicity around its behavior. “Facebook had a philosophy of ‘move fast and break things.’ Now it’s seeing the consequences,” he said. “Ethics and the social sciences need to be part of the [computer] curriculum. We’re trying to hire a philosopher of technology ethics at MIT right now.”

“I don’t think this is going to be effective if it’s voluntary,” said Sen. Brian Schatz, (D-Hawaii). “Coders are not going to wake up and think about the ethics” of what they code. “The only way for an effective counterweight … is to establish in federal law to not harm consumers whose data is collected.”