President Joe Biden’s energetic State of the Union address this week happens to be a convenient news hook for our own “State of the Publication” address: last week, we published the 100th edition of the “Securities” newsletter with “Pivotal Pivots.” Given that we are a weekly publication (and crunching some extremely difficult venture math here), that comes out to roughly two years of science, technology, finance and the human condition. Since I already published best-of-the-year lists for 2022 and 2023, let’s direct our attention to the future of the themes at the heart of “Securities.”
This week, there was no better example of those themes than the news swirling around TikTok, which is as perfect a Rorschach test as one can find at the intersection of technology and foreign affairs. Is TikTok an entertaining app that helps tens of millions of mostly children connect with their peers and learn self-expression in the digital age? Or is it a secret front for Chinese state surveillance and propaganda that’s endangering the mental health and swaying the political opinions of the nation’s youth?
The news this week was the rapid development and momentum behind a divestment bill that is currently wending its way through the legislative process on Capitol Hill. Lux, and particularly Josh Wolfe here, is strongly in favor of the bill. As Josh wrote on Twitter:
Josh is hardly alone, given that many other VCs have similarly endorsed not just the watered-down divestment bill currently before the U.S. House, but earlier legislative attempts to ban TikTok entirely as well. The message of this group is simple, which is one reason why the divestment bill got a unanimous 50-0 vote in the House Energy and Commerce Committee (the committee also passed a parallel bill banning data brokers from exporting American user data to foreign adversaries as well).
The Chinese Communist Party uses legal and extralegal measures to monitor and direct companies, including implanting party cells and requiring adherence to a set of opaque and expansive intelligence and national security laws. TikTok’s ultimate parent company ByteDance is headquartered in Beijing, and is thus covered by these arrangements. TikTok, as one of the most popular social networks for Gen Z alongside Snapchat, provides an alluring and tempting target for CCP influence peddlers and intelligence agents, who see a gusher of data on U.S. citizens plus the ability to shape the perception of impressionable youth by tweaking TikTok’s algorithmic feed. Plus, there’s global precedent: India banned TikTok outright in 2020.
Clearly, the United States would like its citizens to use apps that are owned and operated by American-owned corporations and that follow domestic cultural and ethical norms, particularly around those of the First Amendment. This desire is heightened when an app is used so widely among children, and even further when the app is owned by America’s most aggressive foreign competitor.
The case here for a ban or divestment is, indeed, overwhelming.
But. (And you knew that was coming).
With apologies to Josh, my colleagues and most of the VC industry (at least those who don’t have a personal stake in ByteDance), I remain extraordinarily concerned about Congress’s singular focus on one app and the unique precedent underway of banishing one of America’s most popular platforms for speech. I’m also dismayed that this singular focus has excluded a comprehensive approach to solving the problems surrounding TikTok’s corporate context.
H.R. 7521, the bill that mandates a divestment of foreign ownership from TikTok, specifically defines the term ‘‘foreign adversary controlled application” as an application by “any of” “ByteDance, Ltd.”, “TikTok” and its successors. Just in case we don’t get the point, the preamble of the bill states it bluntly:
To protect the national security of the United States from the threat posed by foreign adversary controlled applications, such as TikTok and any successor application or service and any other application or service developed or provided by ByteDance Ltd. or an entity under the control of ByteDance Ltd. […]
The argument that Chinese intelligence officials can gather American user data through apps is compelling. So then, why is Congress intensely focused on one app to the exclusion of the entire catalog of apps on offer from China-based developers? After all, Shein and Temu are among the most popular shopping apps for American consumers, each of which can track IP and residential addresses and match a consumer to their financial information via credit card. Then there are China-developed video games like Genshin Impact by miHoYo that collectively have millions of American players, which presumably skew toward younger users. The same threats around user data apply equally to all other China-based apps, so let’s divest them all.
Then there’s the wider consideration of foreign ownership. The House bill lays out the criteria for foreign ownership as entities that “directly or indirectly own at least a 20 percent stake.” The threshold matters less to me than the operational principle underlying it: how exactly will companies calculate this particular threshold without greater disclosure requirements mandated by the SEC and other regulatory bodies?
After all, one of the great challenges for analyzing foreign ownership of America’s critical technology is understanding the origin of the dollars that arrive with an investor’s funding wire. The Treasury Department has mechanisms within the Office of Foreign Assets Control (sanctions lists, etc.) focused on anti-money laundering, but the U.S. government lacks any tools or datasets for tracking the dollars flowing throughout the venture ecosystem.
No one truly knows where anyone’s money comes from — not the intelligence community, not the FBI, not the general partners of individual venture funds, no one. If a venture capitalist accepts limited partner funding from a multi-family office, they may know the four families that are the public faces of that office, but perhaps he or she doesn’t realize that most of the funds are actually derived from a fifth family that isn’t disclosed. There is a cottage industry of consultants and firms that do due diligence on this subject, but again, there is no legally mandated dataset compiled by the U.S. government to help with verification. Obviously, the people who most want to launder their money or commit industrial espionage through VC investing are going to be the hardest cases to catch.
If we actually want to decouple from China in the context of critical technologies, then Congress would need to pass a wealth transparency act, indicating where money has flowed from and where it is based. Such tracking by the U.S. government is — rightly — anathema to most U.S. citizens, who have reasonable expectations for privacy from government overreach. So we are left with a serious quandary: we don’t want China-based ownership of our technology or social media platforms, but we also don’t want to develop the wealth surveillance required to actually stop it.
That’s precisely the sort of contradiction that obsesses me. If I had to summarize the past two years of “Securities” looking at quandaries like this at the intersection of technology and politics: they have only gotten more salient, and yet, our political responses to them have deteriorated.
Expanding beyond TikTok to Facebook and Instagram, multiple states including Florida and Oklahoma have recently taken action on bills that would ban young children (up to the age of 16, so not that young) from having access to social media accounts (these states are also trying to prevent access to libraries and books, but I digress). Facebook is nearing its two decade anniversary, and we’re still debating whether children should have access to these technologies (I am, unsurprisingly, opposed to banning Facebook or Instagram for children as well).
Or take AI. The quality of AI technology has surged the past few years, but debates around AI policy remain painfully rudimentary. There are tough tradeoffs and value judgments that need to be made, from safety to copyright and ownership to worker displacement and retraining. Yet, the best lawmakers and regulators can seem to muster is to propose banning certain purposes and otherwise trying to add more general regulatory burdens without figuring out the underlying values that we intend to see enacted (see more in “Reckless Regulators”).
And that leads us back to TikTok: there is a values tradeoff here between freedom of speech and market exchange as well as national security. That’s ultimately what the pluralized “Securities” has always been about: how do we exchange our security of expression with the need to be physically secure through robust national security? On some issues, there’s a positive sum proposal that works, and that is gratifying when we uncover it. Technology, though, has a way of creating dilemmas, and we are utterly failing to take those dilemmas head on and own our own destiny. After two years, “Securities” has a lot more work cut out for it.
Podcast: The three revolutions in astrobiology and the search for extraterrestrial life
Astrobiology has seen a series of revolutions over the past three decades that have completely reinvigorated the field. Scientists who were curious about life and biological organisms across the universe once had to handle the so-called giggle factor: the idea that they were kooky crazies searching for UFOs and little green men. With a dramatic improvement to the quality of our instruments and a torrent of new and better data, that giggle factor is now no laughing matter: we increasingly have the means to make progress here like never before.
My guest today is Adam Frank, the author of The Little Book of Aliens and a professor of astrophysics who is focused on improving our ability to identify biosignatures and technosignatures of life throughout the cosmos. He’s just one contributor to a growing community of scientists reinventing our approach to the search for life, a vitality that is leading to the potential first dedicated satellite focused on the search, the Habitable Worlds Observatory.
Alongside Lux’s scientist-in-residence Sam Arbesman, we talk about the trilogy of revolutions that have brought new vigor to astrobiology, how artificial intelligence is upending the search for life, and what we can also learn about Earth and our climate in searching space for the answers of life.
🔊 Listen to “The three revolutions in astrobiology and the search for extraterrestrial life”
Lux Recommends
- Sam and I have been on a book bender. Sam loved Adrian Tchaikovsky’s novel, The Doors of Eden as well Philip Ball’s new nonfiction work, How Life Works: A User’s Guide to the New Biology. Meanwhile, I have been enjoying Kara Swisher’s Burn Book and Amartya Sen’s Development as Freedom.
- I missed this when we were on hiatus, but Tess Van Stekelenburg recommends the latest announcement from the Arc Institute of its new Evo AI model, which offers one of the largest context windows for biologists yet.
- Given our discussion with Adam Frank on astrobiology, Sam recommends an article on the perfect site for exoplanets that was discovered not so far from us. “Late last year, astronomers discovered a fascinating star system only 100 light-years away from us. Its six sub-Neptune planets circle very close to their host star in mathematically perfect orbits, piquing the interest of scientists searching for alien technology, or technosignatures, which they argue would offer compelling evidence of advanced life beyond Earth.”
- Our producer Chris Gates recommends Saahil Desai’s look and accompanying photographs in The Atlantic on America’s Last Morse-Code Station. “The crew has gotten slightly larger over the years. Its members call themselves the ‘radio squirrels.’ Every Saturday, they beep out maritime news and weather reports, and receive any stray messages. Much of their communication is with the SS Jeremiah O’Brien, a World War II–era ship permanently parked at a San Francisco pier.”
- Sam enjoyed Will Douglas Heaven’s look at the complexities of explaining and understanding what’s happening with large-language models. “The rapid advances in deep learning over the last 10-plus years came more from trial and error than from understanding. Researchers copied what worked for others and tacked on innovations of their own. There are now many different ingredients that can be added to models and a growing cookbook filled with recipes for using them.”
- Finally, Sam and I both enjoyed Derek Thompson’s look at "The Americans Who Need Chaos.” “The researchers came up with a term to describe the motivation behind these all-purpose conspiracy mongers. They called it the ‘need for chaos,’ which they defined as ‘a mindset to gain status” by destroying the established order.’”
That’s it, folks. Have questions, comments, or ideas? This newsletter is sent from my email, so you can just click reply.
