The Mark Zuckerberg that walked into a Senate hearing Tuesday afternoon in a suit and tie to the clicks of several dozen cameras is a very different Mark Zuckerberg than the college sophomore that created a social media platform out of a dorm room. Gone are the hoodies and t-shirts, replaced with a somber blue suit and tired eyes. And just as the social platform changed from a dorm room start-up into a two billion user behemoth, Facebook could very well be at the precipice of another major change.
The Facebook creator and CEO testified before the Senate on April 10 in the wake of a “breach of trust” that had a third-party app collecting user data of up to 87 million users and selling the information to Cambridge Analytica. And while Zuckerberg released his prepared statements ahead of the testimony, which echoed earlier statements on the issues, Zuckerberg also faced questions from each member of the Senate in attendance, addressing concerns from privacy to election interference.
During the course of the hearing, Zuckerberg took responsibility for some of the company’s actions and deflected other topics, saving some questions he couldn’t immediately answer for the written portion of the hearing. Zuckerberg said that the social network is shifting from a reactive approach to how it handles abuse to a proactive one. The Facebook CEO also said that he would support some legislation for social media regulation and privacy laws, in addition to implementing the changes for Europe’s upcoming General Data Protection Regulation (GDPR) laws worldwide.
Senator Richard Blumenthal (D-CT) pointed out that the Global Science Research app terms itself gave the app permission to sell the data, terms that were in violation of Facebook’s policies. Zuckerberg says that those terms should have meant that the app was never approved for the API program.
“Facebook is an idealistic and optimistic company. For most of our existence, we focused on all the good that connecting people can bring,” he said in both the prepared statement and during the hearing. “…But it’s clear now that we didn’t do enough to prevent these tools from being used for harm as well. That goes for fake news, foreign interference in elections, and hate speech, as well as developers and data privacy. We didn’t take a broad enough view of our responsibility, and that was a big mistake. It was my mistake, and I’m sorry. I started Facebook, I run it, and I’m responsible for what happens here.”
The platform has already limited third-party data access, drafted new data policies designed to be more clear, notified affected users, re-designed the security settings menu, and launched investigations into every app that had access to friend data prior to Facebook’s API changes in 2015. Earlier on Tuesday, Facebook announced a new bug bounty to find potential insecurities.
Facebook has a lot at stake — the CEO arrived to around 100 cardboard cutouts of Zuckerberg wearing a shirt saying “fix Fakebook” on the lawn. Facebook says they haven’t seen significant drops in user counts with the #DeleteFacebook movement, but the company has lost $100 billion dollars in stock value since the scandal and Zuckerberg’s own net worth has dropped over $10 billion. The company is also facing lawsuits and some investors are calling for Zuckerberg to step down.
Zuckerberg says Facebook’s approach is now more proactive
When Facebook first started, Zuckerberg explained that the artificial intelligence to monitor the content of posts didn’t exist. Now, the company is shifting towards a more proactive approach, rather than simply reacting to posts flagged by users, Zuckerberg says. For example, Facebook can now remove 99 percent of ISIS posts through AI without not human reviewers. The shift isn’t entirely motivated by Cambridge Analytica, as the platform’s AI tools have already been implemented into several areas.
That capability doesn’t extend to every category, however. Hate speech, for example, is difficult for an AI to detect because of linguistic nuances, slurs, and slang. For what AI can’t detect, Facebook began the process last year of expanding human review staff and is now up to around 15,000 and will reach 20,000 before the end of the year. In the UN, Facebook’s inability to remove hate speech quickly is being questioned in ethnicity-based violence in Myanmar. Zuckerberg says that the company is expanding review staff in the Burmese language, stating that removing hate speech is difficult without a native speaker.
Cambridge Analytica isn’t the first time the social network has apologized over privacy practices — Senator John Thune (R-SD) called the company out on a pattern of lax privacy practices, asking Zuckerberg to clarify how Facebook’s statements now are different than the past apologies.
“It’s impossible to start a company in your dorm room and grow to the size Facebook is now without making mistakes,” Zuckerberg said. He explained that the company would be taking a broader philosophical approach than in the network’s early years, focusing not just on building tools, but on preventing those tools from being abused. “We need to take a more active view in policing the ecosystem and making sure these tools are being used in a way that’s healthy.”
The hearing also looked at Facebook’s business model, with a large majority of the company’s revenue coming from advertising and using user data to deliver those ads to relevant audiences. Zuckerberg says that the network isn’t planning a move to a business model generating revenue by charging users. “There will always be a version of Facebook that’s free,” he said. “It’s our mission to connect people around the world and to do that, [Facebook] has to be free.”
New privacy legislation and opt-in requirement proposed
The hearing also looked at potential legislation for social media regulation — and on the same day, Blumenthal and Senator Ed Markey (D-MA) on Tuesday also introduced the CONSENT Act. The bill, short for the Customer Online Notification for Stopping Edge-provider Network Transgressions, proposes opt-in permission rather than the sometimes hidden opt-out. Zuckerberg also said that the upcoming GDPR laws will also be adopted outside Europe, regardless of if Congress decides to pass similar privacy laws in the U.S.
Senator Orrin Hatch (R-UT) asked the Facebook CEO to offer insight into what legislation would help and what legislation wouldn’t help. Zuckerberg suggested looking at three different points. The first, creating a simple way to understand data use, presented in something easier to understand than a legal document. Zuckerberg also suggested allowing the user to remain in complete control over the data and who sees that information. The Facebook CEO also suggested the lawmakers ensure that any legislation finds a balance between upholding privacy and preventing innovation.
Another senator suggested requiring networks to inform users within 72 hours of a data breach. Zuckerberg said that the suggestion “made sense” but should be discussed in more detail.
The idea of regulating social media wasn’t uniformly supported across members of the Senate. Senator John Neely Kennedy (R-LA) told Zuckerberg that he didn’t want to regulate Facebook, but that he would if he felt regulation was necessary. “Our promised digital utopia, we’ve discovered, has a minefield,” he said. “There are impurities in the Facebook punch bowl that have to be fixed — and I think you can fix them.”
Regulation, whether by the government or the social media company itself, however, tends to bring concern over censorship and ensuring the network isn’t removing posts from one group more than another.
Senator Ted Cruz (R-TX) said that Facebook has blocked a Fox news reporter, dozens of Catholic Pages, and the pro-Donald Trump Diamond and Silk page, and asked the CEO if he knew the political orientation of those 15,000 review staff members. Zuckerberg called Facebook “a platform for all ideas” and said that the company doesn’t ask for political views when hiring. The company has never fired an employee for political views either, he said.
The five-hour session also examined other abuse on the platform, including interference in the U.S. election. Zuckerberg said that, while the company hasn’t found a connection, that the data used in the Internet Research Agency campaigns could possibly have some overlap with the data used by Cambridge Analytica. The new verification process to verify political advertisers using a physical address and government ID will also be applied to administrators of “large” Pages, though Zuckerberg didn’t clarify at what point a Page is considered large.
“I believe deeply in what we’re doing,” Zuckerberg wrote. “And when we address these challenges, I know we’ll look back and view helping people connect and giving more people a voice as a positive force in the world. I realize the issues we’re talking about today aren’t just issues for Facebook and our community — they’re challenges for all of us as Americans.”
