Oh the Humanity
Why You Can't Build Apple with Venture Capital
An ex-Apple designer who went on to startup success once told me, "I wish I could give a workshop for Apple alumni jumping into startups, to help them un-learn The Apple Way." As someone who strives to build products with the craft and quality of Apple, it pains me to admit that The Apple Way can destroy a lot of startups. Which brings us to Humane.
Humane raised $230 Million and spent six years to ship an "Ai Pin," with disastrous results.
The pin shipped with fraction of the features promised at its unveiling eleven months ago, and what features that did ship fail most of the time. It's too heavy, and has atrocious battery life, with users expected to swap batteries every few hours as the device runs uncomfortably hot. Its projector is useless in daylight, and its input system is awkward and slow. For a device meant to replace your phone, you expect an experience faster than just whipping out your phone.
It's safe to say the Ai Pin deserves an exhibit alongside Juicero and Clinkle in the museum of Silicon Valley misfires. It stands out from those flops because, unlike your Clinkles, this isn't the story of a young man given too much money and power without the maturity to wield it. Humane's founders worked on the first iPhone.
I don't have any inside information on what went wrong, but I have theories based on my experiences with venture backed startups and my understanding of The Apple Way of doing things.
The Venture Capital Mindset
A question that keeps coming up is how Humane raised so much money while launching nothing. To get there, we need to break down what Venture Capital is, and how VC firms think about investments.
Rich people and rich organizations (think pension funds and billion dollar college endowments) don't throw all their money in a savings account or the stock market. They have so much money that they can place bets in alternative investments. These are hedge funds, real estate, or venture capital a.k.a. "throw money at a ton of startups."
Investment returns always correlate with risk. Back in 2018, a savings account might earn you a 1% every year, with zero risk that you'll lose money. The stock market might earn you 10%, but carries the risk of a 2007 meltdown. The risk in Venture Capital is downright absurd, with many if not most of your investments losing money. But every so often you find a smash hit that pays back 10,000%.
A respected VC once explained that a third of their investments flop and a third break even, so they need that last one third to be so successful that they both cancel-out the flops and make big returns for investors. That's all to say they play the opposite of moneyball: they aren't looking for safe bets, but a few home runs. You only need to pick one Uber-scale success to make up for picking a dozen Juiceros.
While a normal person would be insane to put their retirement into startups, a VC fund may pool hundreds of millions of dollar— even upwards of a billion dollars— from people and groups so rich that losing $1m feels like a normal person losing $20 in the washing machine. The fund casts a wide net, and every once in a while they find an AirBnB.
While the people who manage these investment firms like to paint themselves as "billionaire scouts" that possess some keen intuition to pick the winners and losers, in reality there's a handful of checkboxes that any mediocre dude can follow.
One big signal is whether the startup founder's had prior success. Once you sell your first startup, VCs knock down your door to fund your next one. Whenever you hear about a stupid startup getting a truckload of funding, it's likely a case of a successful founder's sophomoric follow-up to another winning startup.
While Humane's founders never ran their own startup, one of them (Imran Chaudhri's) did work on the original iPhone. How big a role is up for debate, and most Apple people I talk to think that he takes far too much credit for other people's work. However, Imran's personal website is titled "Select Patented Works, Volume 1: 1995-2017," and links to a thousand patents, his name alongside Steve Jobs. While it makes a lot of people I know cringe, it's enough to get your startup pitch in front of a VC.
Within the pitch, VCs look for "moats." These are the features of the company that make it hard for competitors to copy, or otherwise lock your users into the service. Normal people call these "monopolies," but VCs would rather not say the quiet part aloud.
Humane was pitched as replacing the the phone, so look to the iPhone for the moats. Once users trust a service with their most precious photos, they never want to deal with the hassle of switching. Once developers start developing for your platform, it's hard for even the best competitors (WebOS, Windows Phone) to catch up with that momentum.
Finally, VCs look at investments through the lens of, "Why now?" In 2012 Uber showed that gig work could be the future of employment, so VCs rushed to back DoorDash and Instacart and anything else related to gig work. Humane was founded in 2018, a time when everyone realized screens could be toxic. Within Silicon Valley, parents made nannies sign contracts agreeing to hide their phones from kids. Maybe they imagined phones being stigmatized the same way we stigmatize public smoking.
For all its interesting points, Humane had red flags. Married cofounders are a red flag so big you can see it from outer space. I think VCs cut Humane slack because they were desperate to fund anything during the late 2010's.
Without making this whole post about financial theory, just understand that low interest rates lead to more startup investments. Following the financial meltdown of 2008, the US saw an unprecedented period of low interest rates, and there's even a meme going around where everything insane during the 2010's tech bubble is labeled a "Zero Interest Rate Phenomenon."
In 2018, there was more money flowing into Silicon Valley than there were quality startups to invest in. That's OK for VCs because they actually make money whether or not they pick a winner. They charge "maintenance fees" in the range of 1.5% to 2.5%, so when an endowment writes a VC a check for nine figures, the firm makes millions a year. This meant that Humane didn't have to be a knockout, they just had to be the last single person at the bar at closing time.
Startups Favor Speed
Despite all its quirks, Humane might have worked out if it followed a traditional VC startup formula. Instead, they tried to follow The Apple Way, where 1.0 products need to be so insanely polished as to blow people away.
That approach makes sense for Apple. At minimum, they have a reputation to keep up. Sometimes the Apple Way leads to incredible products, like the first iPhone. Sometimes it doesn't work so well, in the case of the Apple Watch, Imran's final project at the company.
The Apple Watch wasn't a flop, but it did struggle a bit out of the gate. That's expected when you try new things. Before you launch, you live in vacuum, and you operate on faith that your theories will pan out. After a launch, you find some of your theories were right ("Apple Watch is a fitness companion"), and some very wrong ("People will spend $10,000 on a solid gold gadget").
The Apple Way works best when they take an existing product and make it amazing. The best pitch for Apple Watch wasn't "The Rolex of Tech," but rather, "A very fancy FitBit."
It also helps when a product leverages Apple's existing ecosystem, and the goodwill Apple had earned from customers. The Apple Watch connected to the Health app, received messages from your phone, played your favorite music, etc. Apple has a beautiful moat, I'm sure filled with stunning koi fish.
Humane spent five years developing their product in a vacuum. They lacked a FitBit to prove their concept. They had little evidence people want to ditch their phones. They didn't know what form factors users would tolerate. They didn't have normal people telling them battery swaps are dumb.
But the most damaging consequence of their delayed launch was missing the chance to strike while the iron was hot. Humane sounded like a decent idea in 2018, but that same year the iPhone launched its "Screen Time," which has proven a good enough solution for many to curb their screen addiction. In the following years we've watched a decline in the use of social media, which gives me a "nature is healing" vibe. Phone addiction is still a thing, but it feels more like pot than fentanyl.
The AI Pivot
It's obvious that Humane pivoted to AI late in the game, and it's obvious how they got backed into that corner. The formula for a VC backed company goes:
- Raise a little bit of money
- Build something
- Prove it's taking off
- Go to step 1
The more you can show a VC that your company is taking off, the more generous the terms of their investment. Being in a position of power means the difference between giving up 5% or 20% of your company for the exact same money.
This was a lesson lost through the 2010's as money flowed freely and entrepreneurs raised way more than they needed. When interest rates went up in 2022, startups suddenly found it harder to raise money. Investors were suddenly interested in— GASP— profitability.
At this point Humane had raised $130m, which sounds like a lot of money, but they were also growing their headcount to hundreds of employees. This is insane when you haven't validated your product. Knowing typical tech salaries, I'll guess they were burning $30m per year in salaries alone. On top for this, I'm sure they'd want to sell their product at a loss, the same way that Sony and Microsoft sell game consoles to grow their user base. Gotta build that moat.
Humane was in a precarious position for raising more money. Normally, investors expect startups to find an "exit" (IPO or get acquired) somewhere between five and eight years from its inception. That's when investors get their money back with profit. According to VC schedule, Humane should already have a product on the market with a growing user base, and at least showing a path to profitability.
To make matters worse, Humane needed an investor to write a big check. One reason you only ask for less money in your early years is that early-stage investors are comfortable losing a few million here and there. The late-stage investors that write nine figure checks look for safer bets, and they can get by with mere 5× returns.
So Humane needed to ship something to look like a later stage company, but as we now know, their hardware sucked. They decided to latch on to AI, which was in the midst of a huge hype cycle. They shifted messaging away from the projector, and in May 2023, Imran delivered a Ted Talk with an obviously rigged demo that focused on AI.
I tweeted my theory about an AI pivot in November 2023, and it found its way to the official Humane fan Discord. (Yes, Humane setup a fan Discord before announcing any sort of product. They also ran a podcast dedicated to how awesome they were.)
Bethany replied:
Now when someone says "you're wrong" without telling you how you're wrong, you're probably more right than wrong. Bethany tries to hand wave things away by mentioning "the pitch and vision," but… come on.
I'm sure their 2018 pitch was about a post-iPhone device. I'm sure it mentioned a Siri-like assistant. There was zero chance that it involved ChatGPT in 2018, because ChatGPT launched in 2022.
If it's true they always envisioned an AI-first experience, and they spent six years of their lives only to ship this, well, that's way more embarrassing than a last-minute AI pivot.
The Power of Failing Fast
The problems with Humane began at their seed stage. They should have tested the idea faster, and rejected it right away. I get that hardware is different than software, and there's no way Humane could have shipped their projector in a matter of months, or years, but it should have been obvious from early prototypes that this product had a minuscule chance of working in the real world. We've already seen the concept debunked.
Yet I tried to withhold judgement after that Ted Talk. I thought, "Maybe they found a solution to projecting in daylight without getting really hot and egregiously depleting the battery." They didn't. They patched over it. Short battery life? Battery swaps! Doesn't work in daylight? Uh… don't use it in daylight!
For all we know, maybe they did build prototypes but convinced themselves it would be solved. These are certain types of problems you can solve at the scale of Apple, but intractable as a startup. Scaling a startup is about deferring those moonshot bets until your company has the momentum to compete at that scale.
There is absolutely nothing wrong with pitching one idea to investors and going in a totally different direction. There's a saying that early-stage investors back the team, not the idea. Twitter started as a podcast platform, and pivoted to social media after getting sherlocked by Apple. Slack started as— I kid you not— an MMORPG.
Humane could have kept true to the essence of their product, detaching from our phones, while not having to fight an unwinnable battle. What about an e-ink phone? Imagine a device that gives you weeks of battery life, but too low-fi to leave you vulnerable to addictive apps. Maybe that idea sucks, but I'd try a hundred others before going all in on the doomed laser projector
Perhaps we can blame the early commitment on raising too much money, too early. I wish more entrepreneurs knew that you don't need to raise money right away. The co-founders must have walked away from Apple with enough money to diddle on silly concepts for a few years without pressure from investors. The moment they accepted their first check, the clock was ticking. They had eight years to build the next Apple.
It's that funding that will be Humane's undoing. Before the Ai Pin launch, they convinced investors the company is worth $850 million. When they go back for their next round, they need to argue with a straight face that the company is now worth $1.7 Billion. Good luck with that.
It's very likely that the next time Humane raises money, they'll have to value the company less than what they told their last group of investors. These "Down Rounds" screw recent investors and employees, as the stock that they bought is worth less than what they paid for it. It also has a chilling effect on longtime employees. Do not be surprised by an exodus.
The Hubris
I don't write essays about startups failing because it feels like punching down. I wouldn't want anyone doing it to me. In fact, my own apps have been plagiarized more than a few times, and I'll usually bite my tongue rather than trash talk the developers. It's just tacky and a waste of energy.
If I had to give one bit of PR advice to a startup, it's to be humble. I learned this from Twitter's early early years, when users somehow tolerated weekly service outages. I think users forgave Twitter for all its mistakes because Ev Williams is a good guy who cultivated a humble culture. Nobody publicly trash talked Facebook. Even flops like Google Buzz were relegated to a few chuckles on internal mailing lists. I didn't know anyone at the company driven by destroying the other guys.
Humane's follies were amplified by overconfidence, as if they thought they were still working at Apple. When Apple launches a thing, they tell you it's the best thing in the world. More often than not, they're right. People are attracted to confidence, so I get why it works in Apple's favor, but if you come for the king you best not miss.
If you aren't 100% confident your product will work (which is to say, every startup), hype and overconfidence are self sabotage. People forgive of underdog startup that swings and misses, but when an overconfident dude trash talks the competition and falls flat on his face, everyone laughs.
Earlier that year Humane released a trailer that vaguely teased something like a projector system. (Weird choice for an AI centric product?) The trailer felt like a battle cry in the same vein as Apple's 1984 commercial. Don't get me wrong, Apple's commercial was awesome, but maybe it was that hubris that Steve Jobs got fired from Apple a year later?
When Steve Jobs returned, he pushed a culture where Apple pre-announces nothing. Maybe he knew that if you let your customer's imagination run wild, you have no hope of meeting their expectations.
Only hubris can explain Humane's dumbest product decision: a mandatory $24/mo subscription that uses a separate phone number. Set aside the money, nobody wants to manage two numbers! I have two email addresses and regret it whenever I fill out a form.
I can picture the Humane product meeting in my head. "When we released the iPhone, we didn't design it as an accessory for your current phone. It had to be so amazing that users will be ready to go all-in. We cannot compromise Ai Pin by making it an accessory. We must burn the ships."
The flaw in this reasoning is that everything before the iPhone sucked. People were happy to switch to a smartphone, aside from $600 price tag which was hard to swallow.
Today we are so reliant on smart phones that abandoning them would upend our lives. My brokerage account uses a bespoke authentication app, so without my phone and that very specific app, I can't access that money. That would be solved if my brokerage firm ported their authenticator to the Ai Pin, but that will never, ever happen.
It's a credit to the ridiculous momentum of the first iPhone that Apple convinced so many developers to learn a completely new set of skills, and those developers have now spent 16 years in this ecosystem. Why waste time and energy learning a new platform when it doesn't look like there's going to be a market for their apps?
A better strategy for AI Pin 1.0 would have been to release it as an accessory that pairs with your iPhone. Drop the cell connection to improve battery life and bring down costs. If users bring their own carrier, there's fewer complaints about dropped connections.
While we're cutting features, drop the camera. How many people want to take photos that they can't frame ahead of time? Save money on parts, and use that space to fit more battery. Lower costs opens things up to more users, which entice third-party developers to get into the ground floor of the next big thing.
Instead, Humane tried to cosplay as Apple, adopting its "My way or the highway" attitude without any of Apple's leverage that makes it work.
The Danger of Nostalgia
Ken Kocienda is a microcosm of Humane's obsession with the past. Ken worked on the original iPhone virtual keyboard, and has reminded everyone within earshot for the last two decades. Since leaving Apple, he decided to make "Apple is over" his personal brand with all the angst of a dude who got dumped and insists, "I dumped her!"
After tweeting so many strong opinions on design, engineering, and product, Ken launched the first product of his own in 2020, Upspell. It failed to catch the world on fire, and the following year he was hired by Humane to lead product engineering.
I'm not here to make fun of UpSpell. It's a perfectly fine game. But Ken made a very big deal about how he wrote it with Objective-C, and it's so much better than Swift. He published a video about it, which boils down to, "I already know Objective-C and Swift is too complicated." He presents a false dichotomy where you can either work on Swift or focus on building great user experiences.
Ken's opinions fly in the face of most iPhone developers who rejoiced in the move to Swift. Sure, it requires learning new things, but Swift rewards you with fewer bugs and greater productivity. I spent eight years writing Objective-C before moving to Swift, and I've never regretted the switch. I now build features in hours instead of days, and that gives me more time to deliver great user experiences.
There's an irony in Ken's video. He argues that apps are about great experiences and not technology, but instead of making a video about how awesome UpSpell is, he made a promotional video about technology.
The Humane projection system looks really cool, but the experience sucks. The AI experience sounds cool on paper, but the results suck. For all of Humane's talk about freeing users from technology, they shipped technology in search of a purpose.
I blame the team's nostalgia. They clearly want to recreate the Apple from 2007, but that's impossible under venture capital constraints and without the momentum of Apple. Contrary to what Imran, Ken, and I'm sure many others at Humane believe, the iPhone didn't begin with their work in the 2000's on Project Purple. It began in 1976 with the Apple computer, and the decades of goodwill it built up in consumers. The project was spearheaded by a guy ready to waste billions in iPod revenue if it helped achieve his vision, and he answered to nobody. It came together at the perfect point in time, when everyone knew the power of the Internet, but there wasn't a way to carry the whole experience in your pocket. You can't replicate all these factors in a few years, no matter how much money a VC throws at you.
I love Apple's products, and that's why I build my own products outside the venture capital system. I also advise and invest in companies that are compatible with venture capital, because it can actually work out fine if you're prepared for the faustian deal. For my own projects, I'm happy building polished products that will never be worth billions of dollars. But what would I do if I wanted to build the next iPhone?
Imran left Apple in 2017, and his departure email suggested that Apple's best days were behind it. Seven years later, it launched the Vision Pro, a gargantuan effort on par with the original iPhone. We still don't know where it will go, but the launch itself makes it clear Apple is still full of passionate, brilliant people ready to tackle the next big thing.
If there's a tragic lesson to learn from Humane, it's that if you spend all of your time dwelling on the past, you'll miss your chance to shape the future sitting right there in front of you.