Betting on fake humans
Welcome to Cautious Optimism, a newsletter on tech, business and power.
Thursday. The government is open again, perhaps leaving us in the lurch regarding October economic data. Cisco is nearly back to its March, 2000 peak share price, which seems fitting given that the old high point was set during the dotcom bubble. Speaking of bubbles, it appears that everyone and their dog intends to get into prediction markets. With the line already blurry between sports betting and prediction markets, I wonder how many individual players in the game will survive. Waymo is now offering rides on freeways and at more airports. Drip, drip, drip the future is slowly coming into view. To work! — Alex
📈 Trending Up: No shit … Starbucks’ labor woes … oh so that’s why he did that … PC competition … the Qwen brand … Microsoft’s access to OpenAI tech … China-Canada relations … consumer debt …
📉 Trending Down: World — and domestic — religiosity … willing compliance … the First Amendment … public safety in India …
Things That Matter
You can bet on humanoid robots: You, a regular human, cannot invest in Figure or 1X or Apptronik. Those companies are private, and unless you are an LP in a venture fund or wrote very early angel checks, you aren’t getting in. But the hoi polloi have several other options to consider if they are big on two-legged robots taking on a fat share of GDP.
Tesla is the obvious option, thanks to its Optimus robot that the car company’s CEO is very bullish about. There are other investment candidates, and it seems that Xpeng’s recently viral robotic progress — videos of Xpeng’s human-styled humanoid robot walking, and proving its synthetic identity, went viral — is driving retail attention to its shares.
Shares of the Chinese company Xpeng are up about 25% over the last month, driven by investors buying in on hopes of securing a bit of upside from humanoid robotics. It’s worth keeping in mind that Xpeng is, today, nearly entirely comprised of EV revenues. Some $2.36 billion of its $2.55 billion worth of Q2 2025 revenue — Q3 earnings come on the 17th — came from selling cars in the period, vehicles that naturally come with some self-driving technology built in. (Xpeng expects mass-produced L4 self-driving cars in 2026, with “pilot robotaxi services launching in selected regions.”)
During Q2 earnings, Xpeng’s earnings call mentioned its upcoming robots as a tease, but the conversation was nearly entirely focused on cars. Who wants to bet that that changes in a few days?
Humanoid robots, much like SOTA AI models and advanced chipmaking capabilities, are becoming a new front in international competition. Hence why a Russian humanoid robot embarrassingly falling to its face during an unveiling was so damning. The heart of the former Soviet Union is far from achieving Sputnik-style technology advances.
The space economy is accelerating: Earlier this week, I sat down with Orbital Operations’ CEO, Ben Schleuniger, to discuss his startup’s progress in building space tugs. Orbital Ops’ future satellites will use an innovative fueling system to deliver much more thrust in orbit than technologies like electric propulsion can. (It’s a really cool company, and you can check the whole chat here.)
Stepping back from his company, however, I asked Schleuniger about the state of the space economy, asking if it is accelerating as quickly as it appears from my vantage point (quote modestly condensed):
It it it feels like we’ve entered a new era. And I will say a lot of it is coming from the government’s want — not even willingness — but want to work with startups. […]
[Previously, the government would tell startups] if you get all the way to the end, maybe you can partner with a prime. [But] from from what we’re feeling from the top and where where the government’s going, they don’t want to work with the primes anymore and the way primes have done things. Right?
And I’m not shy to say that. They are really pushing to bring innovation in. They wanna see [Apex Space] and [K2] and [Albedo Space succeed], and they want to get that startup mentality and bring that innovation to the forefront because it is starting to be a little bit more of a contested environment, a little more of a crowded environment on the commercial side of things. And we’re realizing that our in space logistics and technologies needs to be pushed and catch up, especially if Starship comes online and all these other launch vehicles come online. So it is 100% [different].
Hell yes. Government demand mixed with private-market purchases is just what we’d want to see from a nation serious about investing in and improving its space muscle. The momentum that Schleuniger discussed is showing up elsewhere in space-land. Firefly Aerospace — our IPO coverage from earlier this year is here — saw its shares soar this week after it reported Q3 earnings that beat expectations.
The launch company is still deeply unprofitable, but will see between $150 million and $158 million worth of top line this year, up from its prior estimates of $133 million to $145 million. SpaceX is great, and gives the United States a clear global edge on rocket launch capacity. But Stoke and Firefly and Blue Origin and Rocket Lab and Astra Space want a piece of the pie. And the more the nation bristles with rockets getting ready to bounce from the ground, the happier I am.
As with Xpeng, it’s cool that retail can get into space launch companies, even if the best-known names remain private. An odd dynamic; in a sense, retail investors are therefore more venture capitalists than the private capital backing SpaceX, which is both larger and more derisked than its smaller, public rivals.
Maybe the Launch Video Era was a mistake
As startups scale their revenues faster than ever before — see this list of $100 million revenue run rate announcements we covered yesterday — there’s increased attention paid to early growth. While this riff from an a16z investor about how much revenue, and how quickly it takes for an early-stage startup to stand out garnered mockery, the point was illustrative; investors want to back momentum, and so the earlier you can show that in dollar terms, the better.
Enter the Launch Video Era, of which Cluely is a leading light. Its early efforts to get attention by bragging about cheating on Amazon’s interview process, dropping out, and then posting an overwrought video announcing its funding round were great at driving attention to the upstart company. Hedra (video), Wispr (video), Atlgo (video), and others are aping the approach to generating attention (to varying degrees of cringe).
The trend is simple to understand: If early revenue growth is critical to raising capital, then you want as much early revenue growth as possible. You can’t do that if no one knows that your company exists. So, you need early attention growth to drive early revenue growth. Therefore, the more early attention you can get, the higher your chances are at quick early growth. And therefore a fat check from a large AUM investor that can afford to take a shot on you. And then you repeat the feat until you become Salesforce, I guess.
I asked a few investors about the trend the other day on TWiST. To my surprise, Menlo Ventures’ Deedy Das wasn’t bothered, saying that he was more concerned when “the product does not follow [the] marketing,” and that “the fact that you’re doing viral distribution campaigns is immaterial” to him. Fair enough.
There are limits to the trend. Cluely admitted as much during TechCrunch Disrupt this year, telling my former journalistic home that driving early attention to a product that might not have been fully baked may have been an error. That isn’t stopping startups from generating attention — even if it’s in the form of annoyance — as an almost moral good in and of itself.
Enter Chad, a new developer IDE that allows users to fuck around more easily while AI writes their code. Sure, developers have long been known to goof around when code was compiling, but in today’s vibe coding market, devs are waiting for code to be written. That means more downtime, so why not watch TikToks while Chad does your work? (TechCrunch points out that the team behind Chad was great at getting attention for their prior project, Party Round.)
Jordi Hays of TBPN fame is not impressed by Chad:
In 2025, rage baiting has become a product strategy. Cluely started as an app for cheating on coding interviews. Chad IDE’s only known differentiation from the other hundred AI native IDEs is that you can gamble and swipe on dating apps in it. The rage bait is sitting at the product level now.
It’s becoming clear that while rage bait might occasionally work as a marketing strategy, it really should not be employed as a product strategy. Running a successful VC-backed company requires you to build a coalition of people that want to see you win. Getting media, investors, talent, and customers on your side is not an easy task. Rage baiting (whether at the marketing level or product level) is the most effective way to get people (who could be potential investors, customers, or team members) to actively pray for your downfall.
I agree with all of that, but what’s odd is that so long as companies that are good at commanding attention raise funds, startups will continue to try to command attention. There’s a bit of Silicon Valley muscle memory at play here; founders that can hack media, or hack attention, are sometimes able to grow their companies more quickly than their peers. Aaron Levie has been doing this for years at Box, leveraging his popular personality to generate a steady stream of attention for his enterprise productivity service, which would otherwise garner as much public attention as, say, Smartsheet.
VCs like founders who can bend the system to their benefit. Founders, almost by definition, don’t win by playing by the existing market rules. The current Launch Video-cum-Ragebait trend is therefore a natural outgrowth of that incentive structure. It’s just become institutionalized and commodified to the point that it’s probably now a negative signal to the savvy. It will take time to cycle out of prominence, however, given that inertia is one hell of a drug.