Digital dystopia

The origins of our digital dystopia

By DEREK ROBERTSON 

Congress’ decision to force a sale of TikTok is a radical break from business as usual with the world’s tech giants, a targeted crackdown on one company that has left analysts and historians guessing as to what might come next.

You might also be wondering whose fault this all is. How could an app with close ties to China’s authoritarian government end up on 170 million Americans’ phones, with so little government leverage that a forced sale looks like the best option?

That’s an easier question to answer: It’s the internet’s fault, and America’s.

If policymakers want to do something more than picking off bad actors ad hoc — to actually make the still-miraculous public service that is the internet a better, more secure place — they need to understand how those bad actors have taken advantage of what was supposed to be an open, utopian, collaborative project.

Underneath the TikTok story (and the Meta story, and all the anxiety-provoking platforms that worry policymakers) is a throughline that runs straight back to the dawn of the internet, and the American-built, U.S.-funded infrastructure that underlies it.

Access to early networks was open to anyone with the hardware to connect, promoting a consensus-driven network of user-developed features like email, early computer games, and the standards that allow for the seamless flow of data.

It’s a strikingly different environment from the closed, data-extracting, app-based systems through which most people access the internet today, but its architecture still, at a basic level, makes the whole thing work.

This all has had huge benefits as the internet, the Web, and then the mobile ecosystem grew. Social media platforms drew non-techies onto the internet, allowing for novel forms of collaboration and activism. A vast constellation of new businesses took advantage of the opportunities provided by e-commerce. The smartphone with its app-store ecosystem utterly transformed Americans’ relationships with the world around them, in many cases for the better.

But this has also, as the TikTok saga shows, had real costs.

That openness was necessary when a barebones internet was crying out for development from hobbyists and engineers in its earliest days. It was exhilarating during the years it was dominated by forum-crawling tech enthusiasts. Now, it’s curdled into what can with little exaggeration be described as a geopolitical nightmare, with the world’s main channel for media consumption overrun with abuse, scams, garbage, espionage, and worse.

The internet’s first engineers, working in the 1960s and 1970s under a contract from the Department of Defense’s Advanced Research Projects Agency, needed to find ways to network refrigerator-sized computers using cross-country phone lines. Bob Kahn, the engineer and networking expert, TCP/IP co-inventor, and coiner of the phrase “open architecture,” was a theorist who steered the early internet toward openness in the interest of making it as large, accessible and cooperative as possible.

It all reflected the optimism of the early computer era. “Life will be happier for the on-line individual because the people with whom one interacts most strongly will be selected more by commonality of interests and goals than by accidents of proximity,” wrote network pioneers J.C.R. Licklider and Robert Taylor in a 1968 paper.

Some of what these developers wrote was remarkably prescient. As Katie Hafner and Matthew Lyon cite in their 1996 history of the internet’s birth “Where Wizards Stay Up Late,” Licklider described a networked citizenry “informed about, and interested in, and involved in the process of government,” and a “political process… [that] would essentially be a giant teleconference, and a campaign would be a months-long series of communications among candidates, propagandists, commentators, political action groups, and voters.”

Pretty good for a prediction made before ”Pong” was released. But tech visionaries have a long history of identifying technology’s transformative potential without quite clocking how that transformation will play out in everyday life, and the internet’s architects were no exception.

The good news is that their dream, more or less, came true: The internet now comprises a totalizing, worldwide network of individuals who may adapt it to their will without friction or interference.

The bad news is that their dream, more or less, came true.

The negative side effects of the internet’s free-for-all are well-documented, especially as artificial intelligence models endow any computer user with power that once only would have been available to a master programmer or global corporation. Less understood is how the very fact that anyone can build, upload, and monetize a popular app — harvesting massive amounts of sensitive data from their users — flies utterly in the face of how any other invention would be built or regulated if it didn’t share the internet’s utopian, Promethean origins.

Which leads us back to TikTok. Years after the internet’s foundations were laid, another massive, nigh-impenetrable layer was added to its “stack”: The mobile operating systems, social media platforms and cloud hosts that use the open internet as an almost invisible backbone for the sophisticated large-scale services they provide. Meta, TikTok, Elon Musk’s X, Apple’s iOS and Google’s Android platform all rely on the free flow of data from users, enabled by the basic internet protocol, to fund the operations of increasingly world-bestriding corporations.

The wonder and terror of the internet is that anyone could theoretically build a machine akin to what TikTok has, as long as they can convince users to accept cookies and agree to their terms of service. (And, of course, raise a lot of capital for server maintenance.) Lest we forget, America’s own titan of data collection and app-based networking was founded in a Harvard undergraduate dorm room.

Given the vastly larger stakes for privacy, security and even democracy now, activists and wonks have long recommended that government step in and regulate the internet differently — in a manner akin to a utility, or the banking system, or aviation, all of which are economically crucial networks that eventually yielded to public oversight.

Critics and entrepreneurs have proposed various means of doing this, like Jaron Lanier and E. Glen Weyl’s proposal for “data dignity,” or the decentralized nature of X competitor Mastodon and other “federated” social media apps, or even the promise to replace the internet’s fundamental layer with a blockchain. Each idea has its own pros and cons, but each has real practical shortcomings given how much has been built on the improbable, wobbling foundation of the current internet.

In the particular case of TikTok, it took a national-security argument about a Chinese-owned app for Congress to finally take action against the unaccountable, behind-the-scenes data collection and algorithmic manipulation that are common practice for American companies. The European Union is far ahead of us on that count.

The politicians who are outraged that this has been allowed to go on for so long overlook the reality that TikTok’s free access to U.S. data is a direct consequence of how the internet was developed — evidence that it’s working exactly as planned, if not exactly intended.

One of Europe’s top digital legislators is pushing for more conciliation with the industry.

Axel Voss, a German European Parliament member who worked on landmark regulations including the Global Data Privacy Regulation and the European Union’s Artificial Intelligence Act, said tech legislation needs to be both clear and flexible to allow more room for competitiveness. He told POLITICO’s Gian Volpicelli (for Pros) that institutions need to be more nimble, and respond faster to technologies that pose regulatory challenges.

“After three months, we should have a solution to problems that might appear suddenly,” Voss said, giving the example of AI. He said if it takes the EU three or four years instead, “all the other regions in the world will have already developed it while we are sitting here.”

Voss also warned of the danger of false information and extreme content spreading in virtual worlds: “I do not want to see a dark metaverse controlled by the Mafia and organized crime,” he said, unintentionally invoking, as Gian pointed out, the author Neal Stephenson’s “Snow Crash.”

Meta announced it will license its virtual reality operating system to third-party hardware makers, setting up an open digital ecosystem in contrast with Apple’s closed iOS.

In a blog post Monday, Meta said its “new hardware ecosystem” will be powered by “Meta Horizon OS,” a new word for what used to be proprietary software for Meta Quest devices. The company announced three partnerships: with ASUS to produce a high-performance gaming headset, with Lenovo to produce headsets for “productivity, learning, and entertainment,” and with Xbox that will come bundled with that video game console’s gear.

Meta said it’s expanding that openness to the all-important app ecosystem as well, “beginning the process” of merging its official app marketplace with the “App Lab” where developers tinker with new apps. In an Instagram video, Meta founder Mark Zuckerberg said it will “probably take a couple of years” for the new devices to reach the market.