The TikTok Time Bomb

A quick note before we begin: we refreshed the “Securities” newsletter today with a more streamlined template that should look particularly better on desktop. Reply to me if you see anything wrong as we continue to work out the bugs.

Now onto our regularly scheduled programming…

Cherry blossoms, Congress, China and control

The cherry blossoms are blooming in DC early this year, but that flowering sakura of ephemeral beauty is hardly lightening the mood in the corridors of the nation’s power. This week — outside of the on-going meltdown in banking, of course — the primary focus was on TikTok, and specifically the Congressional hearing pitting TikTok CEO Shou Zi Chew against the members of the House Committee on Energy & Commerce.

It had all the ingredients of a surrealist spectacle, and a spectacle it was. It had a loaded title: “TikTok: How Congress Can Safeguard American Data Privacy And Protect Children From Online Harms”. It offered a rare opportunity for bipartisan consensus, given the ease of punching down to a foreign-run company CEO without much of a domestic employment constituency.

And as always, you have the lurid dynamic of mostly aged congressmen pontificating on a technology platform mostly used by teens with a 40-year-old at the helm. Of 52 committee members, only Kat Cammack of Florida (aged 35) and Dan Crenshaw of Texas (aged 39) were younger. Only one congressman on the committee, according to WaPo's David DiMolfetta, even has a verified account on the country’s second most popular social media platform (YouTube is number one based on Pew Research polling).

Ignoring Chew’s vetted obfuscations and evasions as well as the committee’s expected grandstanding, TikTok offers an excellent Rorschach test for many prime themes in tech and politics today. Is the concern about China and foreign influence over youth? Is it about protecting American industry from foreign competition? Is it about data privacy, a perennial issue around social media hardly unique to TikTok? Is it about the zombie-like state that teenagers seem to enter as they start scrolling through their feeds? Is it about youth mental health as their minds drown in spastic, rapid, short-form videos?

TikTok embodies and entwines so many of our society’s ills as well as our completely inadequate responses to them. For all the fury unleashed during Thursday’s hearing (and the hearing was definitely heated), we must remind ourselves that all of our anxieties about TikTok were years in the making and always preventable by a functional legislature. Foreign influence could have been checked, data privacy bolstered, and youth mental health ameliorated with legislation. The same threadbare conversations are simply being re-argued again.

Nonetheless, political momentum and DC bandwidth has coalesced toward making a decision: what to do about TikTok? A ban, a divestiture, a new set of operating rules, a data sovereignty agreement — the possibilities are endless. Our own Josh Wolfe posted on Twitter that “TikTok maybe started innocently or maybe was a brilliant Trojan Horse—there is 0 doubt it is a tool of the CCP to engineer, influence, make susceptible the beliefs and emotions of US citizens while also censoring info they see + openly surveilling keystrokes + biometrics.” He’s in the ban camp.

As is our own Bilal Zuberi live in DC at The Hill & Valley Forum, who tweeted that “It’s pretty amazing that from USA to India to New Zealand, governments and intelligence agencies have figured out that TikTok is spyware, but we can’t seem to find out a way to shut it down? How much of our country are we willing to put at risk for entertainment?”

TikTok is the most brazen and well-known hole in America’s information defenses, but it’s merely the largest of a wide-scale intelligence mesh that encompasses tens of thousands of other apps and tech companies (many but certainly not all of them foreign owned and operated) that suck up and exfiltrate the data of American citizens every single day. TikTok is hardly unique, even if it is the most successful at parlaying its data collection into an addictive social product.

Why do we keep allowing this covert data collection to occur? For more than a decade, large tech companies led by Google and Facebook have normalized the open sharing of private user data for the purpose of personalized targeting of advertising and revenue growth. They have been wildly successful, despite widely-read books like Shoshana Zuboff’s The Age of Surveillance Capitalism and widely-covered whistleblowers like Frances Haugen. State-led initiatives in California like the CCPA and elsewhere have tried to tackle the problem, to relatively limited effect.

Despite 1,802 data compromises in 2022 and report after report of the surveillance possible with our apps and devices, there’s been little political momentum to fix the country’s laws allowing the open sharing of private data. That shouldn’t come as a surprise though — the American public is hardly attuned to the issue, although certainly more and more people are actively taking steps to protect themselves where the government won’t (which leads to Lux’s own thesis around privacy technology and our investments in privacy companies like Cloaked and DuckDuckGo).

Banning TikTok may provide some immediate gratification in a short-form world, but that action shouldn’t be a sui generis performance of legal retribution. Instead, it should be a lighthouse that guides the DC blob to a wider conversation about the extent that tech has leached into our lives, our families and our data, and where exactly the line should be drawn between the commercial sphere and the privacy of our own information.

That debate was already fraught before given the trillion-dollar scale of the digital economy underwritten by our current lack of privacy. With the skyrocketing growth of generative AI and chatbots, that debate will only intensify given the voracious need for AI training data to perfect these products in a competitive marketplace. Plus, the quality of products will widen considerably between people conscious of privacy (and therefore limiting the training data available to their AI systems) and those who freely upload their data to every service.

Less than an economic or policy question, this is a question of values and capabilities. There are tradeoffs between the pace of innovation and the protection of our individual sovereignty to choose how we manage our affairs. There are tradeoffs between the instant ease-of-use that Americans expect of new technology products and privacy-preserving features that require users to think and consent to how their data might be used.

But we need to have that debate on values, since data is fundamentally fungible and also overflowing today. If China doesn’t get data via TikTok, it will get it through dozens of other popular Chinese-owned apps, or even via agreements with other companies, or frankly, just buying it wholesale through the existing data brokering market. If the data is “out there”, it’s discoverable by a nation-state dedicated to seizing it. If we truly desire to protect user information from evil actors, then we actually have to protect it.

That might include strict legal liability for corporate data breaches. That might include the creation of what I’ll dub a “Data Force” as a new military branch dedicated to defending cyber data and retrieving it when necessary. It would certainly include more requirements for user control over how their personal information is used, and explicit power to block the transfer of all data to third-party brokers and other data resellers. It might also include a blanket ban on data brokering in the first place.

TikTok is a test, but it’s one that we only solve if we look beyond its phantasmagoric Rorschach shapes and start to consider the wider landscape of its place in the ecosystem. A ban would be effective in stopping some Chinese data collection, but much like a cherry blossom, the app’s quick bloom and disappearance will simply come once again in new configurations in the years ahead unless more complete action is taken.

Thematic opens up smart democratized finance for all investors

We’ve mentioned Thematic in the newsletter a bit, but it’s a platform to easily build custom thematic market indexes as well as ETFs. Conceived by serial entrepreneur Steven Carpenter and Lux’s own Peter Hébert, Thematic officially launched and announced their fundraise from us this week.

A lot of the attention around democratizing finance got sucked up into the crypto trading world the past few years, but the core of the movement is about helping the unbanked and underbanked move into traditional banking, and then progressively opening all products and layers of the financial system to carefully regulated competition. Better products, lower fees and greater access combine into something special.

Thematic’s goal is to offer a balance between the day-trading frenzy of a platform like Robinhood and the extremely passive and low-signal big ETF funds offered by the likes of Blackrock and State Street. By having individual experts craft tailored thematic ETFs on the Thematic platform, investors can access a greater range of high-quality investment products that capture alpha.

For instance, our own Grace Isford constructed The Lux Web3 Infrastructure Index last year to capture blockchain networks and companies building on and adjacent to the Web3 world. It’s up 37.72% the last three months (and, of course, past returns do not equal future performance). Be sure to check out other indexes from Lenny Rachitsky on SaaS and Howard Lindzon’s DROP DEAD index.

The perturbations ushering in the next generation of biological research

There’s been a quiet revolution in biology, and it’s a quintessential example of how small changes in science (in this case literally) can lead to large advances.

Over the past decade, scientists have harnessed the exponentially increasing power of next-generation sequencing, single-cell analysis instruments, and other laboratory tools to begin understanding molecular differences between individual cells. Now, those tools are increasingly parallelized, allowing scientists to experiment on tens of thousands of cells with precise, tiny changes to their genetic code in order to glean how cell functionality is derived.

This “perturbation biology” method is rapidly improving our knowledge of the inner workings of the building blocks of life. This week on the “Securities” podcast, I talked with Rahul Satija, an associate professor at New York University and a core member of the New York Genome Center as well as Lux’s own Shaq Vayda about the implications of perturbation bio, how our understanding of cell heterogeneity is improving, and why fundamental molecular biologists will soon have to work with the applied world of large-scale clinical trials.

🔊 Listen to “How exponentials on top of exponentials in single-cell analysis is transforming biology today”

Lux Recommends

We all recommend our portfolio company Runway’s new feature “Text to Video”, which takes a prompt from a user and transforms it into a full-motion video through generative AI. Check out the demo of Gen-2. And also, check out the videos from the AI Film Festival in San Francisco this past week, which our own Lan Jiang, Chris Gates, Scott Rubin and others attended.
Our scientist-in-residence Sam Arbesman is excited about the Vesuvius Challenge, a machine-learning and computer-vision competition to read the Herculaneum Papyri offered by Nat Friedman, Daniel Gross and Dr. Brent Seales. It offers more than $1 million in prizes.
Many of us here were intrigued by a new paper in Current Biology and the write-up in the New York Times on a DNA analysis of Ludwig van Beethoven’s hair, which found that he likely had Hepatitis B from child birth and also ruled out a variety of other potential diseases. It’s a fun puzzle of historical DNA sleuthing.
Sam has been on a reading binge, and recommends three books: a novel on sudden internet fame titled An Absolutely Remarkable Thing by Hank Green, R.F. Kuang’s novel on power and translation, Babel: Or the Necessity of Violence: An Arcane History of the Oxford Translators’ Revolution, and finally, The Long View by Richard Fisher, on how to think long term in a world that constantly reduces our time for thinking.
A fun one from Pete in the WSJ: “MIT Scientists Twist Apart More Than 1,000 Oreos in Search for Perfect Method”. To which Bilal commented, “I must have an MIT scientist for a son” then.
I enjoyed Robert Zaretsky’s earlier work on Albert Camus, “Albert Camus: Elements of a Life”, which does a great job of weaving together writer biography with works of art to arrive at an informed whole.
Finally via Pete, two beautiful maps of African and Mexican rivers by Juan Luis Torres:

Map of Africa designed by Juan Luis Torres

That’s it, folks. Have questions, comments, or ideas? This newsletter is sent from my email, so you can just click reply.

Forcing China’s AI researchers to strive for chip efficiency will ultimately shave America’s lead

In incididunt ad qui nostrud sint ullamco. Irure sint deserunt Lorem id officia dolore non. Anim dolor minim sit dolor et sint aliquip qui est. Ex in tempor laborum laboris dolor laboris ullamco quis. Enim est cupidatat consequat est culpa consequat. Fugiat officia in ea ea laborum sunt Lorem. Anim laborum labore duis ipsum mollit nisi do exercitation. Magna in pariatur anim aute.

In incididunt ad qui nostrud sint ullamco. Irure sint deserunt Lorem id officia dolore non. Anim dolor minim sit dolor et sint aliquip qui est. Ex in tempor laborum laboris dolor laboris ullamco quis. Enim est cupidatat consequat est culpa consequat. Fugiat officia in ea ea laborum sunt Lorem. Anim laborum labore duis ipsum mollit nisi do exercitation. Magna in pariatur anim aute.

‍

Right now, pathbreaking AI foundation models follow an inverse Moore’s law (sometimes quipped “Eroom’s Law”). Each new generation is becoming more and more expensive to train as researchers exponentially increase the number of parameters used and overall model complexity. Sam Altman of OpenAI said that the cost of training GPT-4 was over $100 million, and some AI computational specialists believe that the first $1 billion model is currently or will shortly be developed.

As semiconductor chips rise in complexity, costs come down because transistors are packed more densely on silicon, cutting the cost per transistor during fabrication as well as lowering operational costs for energy and heat dissipation. That miracle of performance is the inverse with AI today. To increase the complexity (and therefore hopefully quality) of an AI model, researchers have attempted to pack in more and more parameters, each one of which demands more computation both for training and for usage. A 1 million parameter model can be trained for a few bucks and run on a $15 Raspberry Pi Zero 2 W, but Google’s PaLM with 540 billion parameters requires full-scale data centers to operate and is estimated to have cost millions of dollars to train.

Admittedly, simply having more parameters isn’t a magic recipe for better AI end performance. One recalls Steve Jobs’s marketing of the so-called “Megahertz Myth” to attempt to persuade the public that headline megahertz numbers weren't the right way to judge the performance of a personal computer. Performance in most fields is a complicated problem to judge, and just adding more inputs doesn't necessarily translate into a better output.

And indeed, there is an efficiency curve underway in AI outside of the leading-edge foundation models from OpenAI and Google. Researchers over the past two years have discovered better training techniques (as well as recipes to bundle these techniques together), developed best practices for spending on reinforcement learning from human feedback (RLHF), and curated better training data to improve model quality even while shaving parameter counts. Far from surpassing $1 billion, training new models that are equally performant might well cost only tens or hundreds of thousands of dollars.

This AI performance envelope between dollars invested and quality of model trained is a huge area of debate for the trajectory of the field (and was the most important theme to emanate from our AI Summit). And it’s absolutely vital to understand, since where the efficiency story ends up will determine the sustained market structure of the AI industry.

If foundation models cost billions of dollars to train, all the value and leverage of AI will accrue and centralize to the big tech companies like Microsoft (through OpenAI), Google and others who have the means and teams to lavish. But if the performance envelope reaches a significantly better dollar-to-quality ratio in the future, that means the whole field opens up to startups and novel experiments, while the leverage of the big tech companies would be much reduced.

The U.S. right now is parallelizing both approaches toward AI. Big tech is hurling billions of dollars on the field, while startups are exploring and developing more efficient models given their relatively meagre resources and limited access to Nvidia’s flagship chip, the H100. Talent — on balance — is heading as it typically does to big tech. Why work on efficiency when a big tech behemoth has money to burn on theoretical ideas emanating from university AI labs?

Turn to China now, and the story is in a very different place. The Biden administration has placed a series of export controls on advanced computation and semiconductor fabrication to China, including blocking the export of Nvidia’s H100 and A100 chips. Nvidia responded in March by offering tweaked versions that comply with U.S. export controls (with what it dubs the H800 chip), but the general policy remains in force: prevent China from acquiring the best chips that Earth produces.

Without access to the highest-performance chips, China is limited in the work it can do on the cutting-edge frontiers of AI development. Without more chips (and in the future, the next generations of GPUs), it won’t have the competitive compute power to push the AI field to its limits like American companies. That leaves China with the only other path available, which is to follow the parallel course for improving AI through efficiency.

For those looking to prevent the decline of American economic power, this is an alarming development. Model efficiency is what will ultimately allow foundation models to be preloaded onto our devices and open up the consumer market to cheap and rapid AI interactions. Whoever builds an advantage in model efficiency will open up a range of applications that remain impractical or too expensive for the most complex AI models.

Given U.S. export controls, China is now (by assumption, and yes, it’s a big assumption) putting its entire weight behind building the AI models it can, which are focused on efficiency. Which means that its resources are arrayed for building the platforms to capture end-user applications — the exact opposite goal of American policymakers. It’s a classic result: restricting access to technology forces engineers to be more creative in building their products, the exact intensified creativity that typically leads to the next great startup or scientific breakthrough.

If America was serious about slowing the growth of China’s still-nascent semiconductor market, it really should have taken a page from the Chinese industrial policy handbook and just dumped chips on the market, just as China has done for years from solar panel manufacturing to electronics. Cheaper chips, faster chips, chips so competitive that no domestic manufacturer — even under Beijing direction — could have effectively competed. Instead we are attempting to decouple from the second largest chips market in the world, turning a competitive field where America is the clear leader into a bountiful green field of opportunity for domestic national champions to usurp market share and profits.

There were of course other goals outside of economic growth for restricting China’s access to chips. America is deeply concerned about the country’s AI integration into its military, and it wants to slow the evolution of its autonomous weaponry and intelligence gathering. Export controls do that, but they are likely to come at an extremely exorbitant long-term cost: the loss of leadership in the most important technological development so far this decade. It’s not a trade off I would have built trade policy on.

The life and death of air conditioning

Across six years of working at TechCrunch, no article triggered an avalanche of readership or inbox vitriol quite like Air conditioning is one of the greatest inventions of the 20th Century. It’s also killing the 21st. It was an interview with Eric Dean Wilson, the author of After Cooling, about the complex feedback loops between global climate disruption and the increasing need for air conditioning to sustain life on Earth. The article was read by millions and millions of people, and hundreds of people wrote in with hot air about the importance of their cold air.

A few weeks ago, Bloomberg had a great piece about the rising toll of air conditioning under the similarly scary headline, “A Billion New Air Conditioners Will Save Lives But Cook the Planet.” The numbers, as always, are staggering:

Demand for air conditioners is surging in markets where both incomes and temperatures are rising, populous places like India, China, Indonesia and the Philippines. By one estimate, the world will add 1 billion ACs before the end of the decade. The market is projected to
before 2040. That’s good for measures of public health and economic productivity; it’s unquestionably bad for the climate, and a global agreement to phase out the most harmful coolants could keep the appliances out of reach of many of the people who need them most.

This is a classic feedback loop, where the increasing temperatures of the planet, particularly in South Asia, lead to increased demand for climate resilience tools like air conditioning and climate-adapted housing, leading to further climate change ad infinitum.

And there doesn’t appear to be a way out. Scientists have determined that there is a zone of comfort for humans where we can function at peak performance. Much of the industrialized world happens to sit in these temperate zones, which is one argument for their long-term economic success compared to the tropics. But as climate change intensifies, the world’s temperate zones are expected to narrow dramatically, and some countries like India may have next to no temperate zones left to protect its people. Air conditioning is a necessity (with some even calling it a human right), but greener coolant alternatives are still in their infancy. In other words, cooling innovation needs to heat up in a big way.

Lux on the Farm

Josh Wolfe speaking at Nvidia Auditorium at Stanford. Photo by Chris Gates.

Josh Wolfe gave a talk at Stanford this week as part of the school’s long-running Entrepreneurial Thought Leaders series, talking all things Lux, defense tech and scientific innovation. The
.

Lux Recommends

As Henry Kissinger turns 100, Grace Isford recommends “Henry Kissinger explains how to avoid world war three.” “In his view, the fate of humanity depends on whether America and China can get along. He believes the rapid progress of AI, in particular, leaves them only five-to-ten years to find a way.”
Our scientist-in-residence Sam Arbesman recommends Blindsight by Peter Watts, a first contact, hard science fiction novel that made quite a splash when it was published back in 2006.
Mohammed bin Rashid Al Maktoum, and just how far he has been willing to go to keep his daughter tranquilized and imprisoned. “When the yacht was located, off the Goa coast, Sheikh Mohammed spoke with the Indian Prime Minister, Narendra Modi, and agreed to extradite a Dubai-based arms dealer in exchange for his daughter’s capture. The Indian government deployed boats, helicopters, and a team of armed commandos to storm Nostromo and carry Latifa away.”
In the happier news department of science, Ina Deljkic recommends “’Quite science fiction’: Brain-spine interface system helps man with paralysis walk again.”
Sam recommends Ada Palmer’s article for Microsoft’s AI Anthology, “We are an information revolution species.” “If we pour a precious new elixir into a leaky cup and it leaks, we need to fix the cup, not fear the elixir.”
I love complex international security stories, and few areas are as complex or wild as the international trade in exotic animals. Tad Friend, who generally covers Silicon Valley for The New Yorker, has a great story about an NGO focused on infiltrating and exposing the networks that allow the trade to continue in “Earth League International Hunts the Hunters.” "At times, rhino horn has been worth more than gold—so South African rhinos are often killed with Czech-made rifles sold by Portuguese arms dealers to poachers from Mozambique, who send the horns by courier to Qatar or Vietnam, or have them bundled with elephant ivory in Maputo or Mombasa or Lagos or Luanda and delivered to China via Malaysia or Hong Kong.”
Finally, Sam recommends Brick Technology’s video of “Making A Billion-Year Lego Clock.”