Over the last two newsletters (three, if you include my reply to Google’s “rebuttal” of the Prabhakar Raghavan newsletter), I’ve made the case that while rot economics are responsible for making technology products manifestly worse, this transformation was only possible thanks to the interventions of a managerial class.
These vile management consultants — like Sundar Pichai and Sheryl Sandberg — and their acolytes who exhibit the same destructive traits (in particular Raghavan) are the faithful foot soldiers in the war to ruin technology products, dampen innovation, and destroy otherwise-flourishing companies in the name of short-term market gains that benefit only the investor class. They didn’t build these products, and they certainly don’t use them - all they care about is market share, revenue and organizational power.
Nowhere is this more obvious than Meta, a company with leadership completely removed from any meaningful interaction with their products. Since 2009, Facebook’s core products have reliably become more profitable as they decay, with every founder behind every product that Zuckerberg has acquired — including Instagram, Oculus and WhatsApp — leaving the company and almost immediately talking about how much it sucked.
According to a New York Times piece from 2018, Kevin Systrom, co-founder of Instagram, only chose to quit the company after Zuckerberg became jealous of the app’s success taking the spotlight away from that of Facebook itself. Systrom allegedly didn’t want to leave the company, but felt that Zuckerberg was depriving Instagram of resources, and “now seemed to want Instagram to use its momentum to help the big “blue app’ — an annoying way of describing a situation that feels like a convenient way to reveal that this was a Kara Swisher piece.
Despite Swisher’s bloviating, it took TechCrunch’s Josh Constine to reveal the real reason that Systrom had left. Facebook had replaced Instagram’s VP of Product Kevin Weil with the former VP of Facebook News in May 2018 — a man named Adam Mosseri who would over the next six years absolutely destroy everything that Systrom and his co-founder Michel Krieger had built.
According to Constine’s reporting, Systrom had also clashed with Chris Cox, Facebook’s Chief Product Officer of the time. Constine described Mosseri as a “Zuckerberg loyalist” who was “disappointed that he didn’t get the head of Facebook Gig” that went to Will Cathcart, who now heads up WhatsApp. Over time, Mosseri and Zuckerberg had moved to erode Systrom and Instagram’s independence from Facebook, and eventually it became too much to bear.
Systrom was there to oversee the most damaging change made to Instagram — the introduction of the algorithmic feed in June 2016 that horrified users who feared that they would now not see posts from their friends, a thing that almost immediately happened on both Instagram and Facebook, which made a major change to its newsfeed algorithm in 2015.
A few months later, Instagram would try to clone the functionality of Snapchat — a company that has had quite literally just two profitable quarters since its inception — with the release of “Stories”. This move was illustrative both of the lack of creativity within Instagram, but also of its future direction, with Stories serving as yet another touchpoint for advertisers.
Once Systrom left, Mosseri became the head of Instagram, turning it into one of the most profitable business units in history while destroying its basic functionality as an app that showed you photos and videos from people you chose to follow, doubling down on the algorithm’s ability to interrupt what you want to see with ads or sponsored content.
Since taking over Instagram, Adam Mosseri has — with the direct approval of Mark Zuckerberg — turned the app into a glorified ad network devoid of any ability to innovate, with products like IGTV and Threads (a camera app built to compete with Snapchat, not the social network of the same name built to compete with Twitter) that never found traction, and every change under Mosseri seems to be a direct copy of either Snap or TikTok.
It’s important to remember that Adam Mosseri is not a creator, or an engineer, or a founder, but a designer that found a way into product management, escaping the doldrums of doing actual work into the pantheon of management. And in late 2020, he made arguably the worst change to Instagram yet, launching “Reels,” a 15-second video format for Instagram built to compete with the ascendant growth of the algorithmically-driven TikTok. Reels quickly became the dominant form of content on both Facebook and Instagram, flooding your feed with 15 to 60 second clips that automatically play as you scroll by, each one either engineered or paid to get in the way of things that you actually want to see.
Mosseri is also one of the least-popular tech executives in history. Since taking over Instagram, he’s had to apologize for an update that made Instagram’s feed move sideways, apologize for Instagram censoring pro-Palestinian content, testify before congress about Instagram’s harm to young people, and tell people that Instagram was no longer a photo-sharing app. He’s overseen so many deeply unpopular changes that Kim Kardashian and Kylie Jenner, who between them have over 600 million followers, had to beg him to “stop Instagram from trying to be TikTok,” to which he responded that “more and more of Instagram is gonna become video over time,” and claimed that it had cut back on recommended content, something I think we can all agree was a big fucking lie.
Mosseri, like many of the most powerful men in tech, is a glorified management consultant incapable of creating anything of note. Mosseri has already announced that Threads, Meta’s dollar store clone of Twitter, isn’t for news and politics, making news organizations hesitant to invest in a platform made by a company that has a rich history of screwing over news organizations. Nowhere is Mosseri’s consultant mindset more obvious than in his suggested plan to deal with Instagram’s hundreds of thousands of underage users by creating a “new type of family-centered account in Instagram that would permit Meta to upsell Instagram to children under 13,” a disgusting program that was planned as an alternative to instituting stricter registration procedures according to a lawsuit against Meta filed by the Attorney General of New Mexico.
This is the man running one of the most important tech platforms in the world — a man bereft of morals or qualifications, or even ideas. A walking, talking figurehead that exists only to spout vague platitudes about what social media can or will do as he profits from making it harder to communicate with your friends and family.
While much of the blame for the state of Instagram and Facebook can be laid at the feet of Mark Zuckerberg, it’s important to understand the sheer damage that Adam Mosseri has done to the world. Instagram is now a truly awful product, beset with scammy advertising and an aggressive algorithm that deprives the user of industry, and Mosseri’s only response to the pain and frustration of his users is to tell them that he intends to make Instagram even worse in the future.
This is the management consultant mindset that dominates tech — trapping users in terrible experiences by monopolizing industries, then making their products worse once they know that their users can’t go anywhere else. Yet their massive success only seeks to inspire generations of useless founders, obsessed with creating profitable pain-boxes over useful products. And as we speak, one of the more loathsome one of them all is rising to power.
I am, of course, talking about Sam Altman.
Do you like newsletters? Well, our friends over at Semafor Tech write a twice weekly newsletter on the people, the money and the ideas at the center of the new era. I genuinely like Semafor a lot - they do great journalism, and I think I've cited Reed's scoop about Microsoft's OpenAI investment three or four times now.
You can sign up for the newsletter here.
Failing Upwards
In 2005, Sam Altman, a Stanford dropout, co-founded Loopt, a location-based social networking app that raised over $30 million from tech incubator Y Combinator and VCs like Sequoia and NEA. Seven years later, Altman would flog Loopt to publicly-traded financial services Green Dot for a remarkable $43.4 million, despite the app’s inability to find traction or revenue. Altman got quite rich from the Loopt deal, despite the fact that a group of senior employees urged the board on two separate occasions to fire Altman for “what they described as deceptive and chaotic behavior,” according to the Wall Street Journal.
Altman would almost immediately become a partner at Y Combinator after working there part-time, before being made president by co-founder Paul Graham in 2014. Yet behind the scenes, according to reporting by Elizabeth Dwoskin and Nitasha Tiku of the Washington Post, Altman was well-known, and I quote, “for an absenteeism that rankled his peers and some of the startups he was supposed to nurture.” Altman also double-dipped in Y Combinator startups through Alt Capital, a venture firm founded with his brother Jack, with one source quoted by the Post describing Altman’s tenure as “the school of loose management that is all about prioritizing what’s in it for me.”
Altman became wildly rich during his tenure at Y Combinator, using his cult of personality and masses of connections to make early investments in companies like Gusto, Optimizely (which was acquired in 2018 for $1.16 billion), Patreon, Reddit and Asana. In 2015, he founded OpenAI — at the time a non-profit organization dedicated to building responsible artificial intelligence applications. Yet it’s critical to note that Altman is not and was not ever an engineer or a technologist, he was a figurehead and a fundraiser that was able to convince actual academics and engineers like Durk Kingma, Wojciech Zaremba and Ilya Sutskever to do the actual work while he sent masturbatory emails to Elon Musk, who only ever donated $50 million of the $100 million he claimed he’d invested into OpenAI.
I really want you to remember that Sam Altman was an absentee parent for the first few years of OpenAI’s existence, splitting his efforts in a Musk-esque fashion across multiple other investments and enterprises, like the two other funds he’d built to run within yCombinator.
In 2019, according to reports at the time, Altman would step down from Y Combinator in 2019 “amid a series of changes at the accelerator,” a story that much of the industry press chose not to investigate. Yet the Washington Post’s reporting revealed that Y Combinator founder Paul Graham flew from the United Kingdom to San Francisco to personally give Altman the boot due to Altman continuing to focus on his own personal projects and press.
This is the story of the man who New York Magazine called “The Oppenheimer of Our Age” in a meandering piece that frames Altman’s vagueness about artificial intelligence as some sort of big secret, when I think the truth is far simpler: Sam Altman is yet another fucking management consultant. In a piece published in early 2021, Altman proposed the concept of “Moore’s Law For Everything” — referring to the principle that the number of transistors on an integrated circuit doubles approximately every two years, leading to a linear increase in computing power in the process.
Moore’s Law for Everything is, in essence, a utopian take on the impact of AI, noting that as machines usurp the role of humans in the supply chain, the prices of goods and services (and thus, the cost of living) will go down exponentially.
The problem with this piece is it, like Altman, is a deeply complex bucket of nothing. It’s an extremely verbose screed that says things like “AI will lower the cost of goods and services, because labor is the driving cost at many levels of the supply chain” and suggesting we “tax capital rather than labor” by creating an “American Equity Fund” where companies are forced to give up a percentage of their shares to a nationally-incorporated venture fund, an idea that makes sense if you’re in the business of flogging private companies to public companies.
“Moore’s Law For Everything” is a remarkably telling piece in that it frames Altman’s worldview as one where the only thing that can save mankind is making startups, and those startups should be funded in the way that Altman likes. And this piece feels a lot like everything in Altman’s universe — it never actually connects Moore’s Law to anything, which, I should add, isn’t so much a law as an observation that has long ceased to be relevant as the shrinkage of transistors has slowed in recent years. In many respects, this comparison is eerily prophetic, given the slowdown (and even regression) of improvement we’ve seen in LLM-based services like ChatGPT.
Much like ChatGPT, Altman is capable only of loosely approximating the output requested, because at his core he lacks any of the substance or technical history required to do so. Like ChatGPT, he’s a know-nothing that, through deterministic measures completely detached from the meaning of the underlying ideas, picks the right words to say at the right time.
And this is the man selling the artificial intelligence dream — a salesman capable of superficial connection of ideas in a way that’s initially satisfying, as long as you don’t think about it too much. Altman’s famed “startup playbook,” published in 2015, is full of the kind of obvious-yet-satisfying pablum that you’d expect. He extolled the virtues of being “flexible yet rigid,” advises that you “talk to your users and watch them use your product” (which is an exact quote), and try to “improve your product 5% every week.” These are the kind of things that are very useful to an early-20s founder and impressive to a mid-50s white venture capitalist that doesn’t remember the last time they worked a job that wasn’t ten hours a week of investing in Chuntly, the SaaS for Dog Breeders. They’re the tech equivalent of a “Live, Laugh, Love” bumper sticker.
It does, at one point, betray Altman’s real mindset — that the only universal job description of a CEO is to make sure the company wins.
Altman’s material contributions to OpenAI are hard to nail down. While it’s unfair to judge someone entirely by their emails, those that I can find (such as this one from Elon Musk’s ongoing lawsuit against OpenAI, found on page 40) feel like they could be from any other managerial huckster, and they feel just as specious as the ones sent by Elon Musk that were recently revealed in OpenAI’s response to said lawsuit.
Altman blathers on about governance structures and how OpenAI needs to “create the first general AI and use it for individual empowerment…” which he defines as “the distributed version of the future that seems the safest.” Musk and Altman are similar creatures — managers wearing engineering costumes, and both are credited as having “expertise in AI” without actually appearing to have written a single line of code in over a decade.
In Altman’s case, I can find little evidence of actual work and a great deal more of a guy who multiple people have tried to fire. We still don’t know the full story behind why Sam Altman was briefly fired from OpenAI in November of last year, but we do know that in doing so, Altman was able to fill OpenAI’s board with his friends and business parties, like Fidji Simo, the CEO of Instacart, a company that Altman invested in, who was previously the head of the Facebook App from 2019 to 2021, also known as “the years that Facebook got much, much worse.”
From what I can tell, Altman has, like Raghavan — the villain of the last two newsletters — failed upwards. Except unlike Raghavan, Altman doesn't appear to have contributed much to the world at all other than money movement and, well, distractions.
He was a constant source of frustration at Loopt due to his pursuit of side projects, with the Wall Street Journal reporting late last year that Altman once diverted engineers to work on an unnamed gay dating app. As I previously noted, Altman was fired from Y Combinator for his absenteeism, and I quote, “reputation for favoring personal priorities over official duties,” and the Wall Street Journal reports that by early 2018 — a year before he was fired — Altman was “barely present at Y Combinator’s headquarter,” instead spending more time at OpenAI, much to the chagrin of other longtime partners at Y Combinator, who “began losing faith in him as a leader.”
The Journal’s piece does reveal a little more about why Altman was fired as CEO of OpenAI, and it’s partly because Sam Altman is a pretty atrocious manager. Founding and now-former OpenAI board member Ilya Sutskever described to the board, when calling for Altman’s removal, “a long-running pattern of Altman’s tendency to pit employees against one another or promise resources and responsibilities to two different executives at the same time, yielding conflicts.” More-worryingly, the Journal reports that other members of the board had heard similar concerns from senior OpenAI executives.
They also feared that Altman would use his influence in Silicon Valley once fired, something that almost immediately came true when Sam called Brian Chesky, CEO of Airbnb, who then called Satya Nadella, CEO of Microsoft, which sparked a chain of events that restored Altman as CEO of OpenAI and led to the removal of the non-believers from the board entirely.
This is Silicon Valley’s king. A lobbyist roleplaying as a founder. A diplomat masquerading as a technologist. A charming, capricious, abusive and untrustworthy man that has proven time and time again that his only reliable trait is that whatever happens must benefit Sam Altman.
This also explains why so little of Sam Altman’s promises about AI make sense, and why OpenAI has been so unashamed in steamrolling and plagiarizing the entire world. Altman has created nothing other than wealth for himself and his closest allies, helping elevate and protect existing power structures and the ideologies of men like Microsoft CEO Satya Nadella and LinkedIn founder and career manager Reid Hoffman.
And it all kind of makes sense why generative AI doesn’t really help anyone other than those who want to sell something. When the center of attention at a company isn’t really on the product but the idea of what the product could do, very little about a company’s culture is focused on building useful things for real people. When leadership is dominated by managers that haven’t touched a line of code in decades, nobody steering the ship has the ability to judge whether software is good, or useful, or valuable.
This is the direct result of Silicon Valley’s corruption by the managerial sect. While Prabhakar Raghavan may be a decorated computer scientist and academic, he arguably oversaw the destruction of Yahoo, formerly one of the web’s most dominant search engines, and failed upwards into a managerial role that allowed him to take over — and now arguably ruin — Google’s search product, chasing away Ben Gomes, a man responsible for actually building things. Adam Mosseri was and always will be a manager making calls about products he has had no hand in building, and has architected the outright destruction of a social network used by billions of people.
And Sam Altman, a career failure famous for making himself rich and popular and upsetting those he works with, is on course to become the most toxic manager of them all. If left unchecked, OpenAI will perpetuate one of the largest thefts in history, looting the internet and using it to train models that have yet to prove their necessity other than as a symbol that Silicon Valley can still innovate.
Yet because Altman — like every manager — is so thoroughly divorced from actual production, he’s only succeeded in generating unsustainable hype and making vague promises that the people who do the actual work at OpenAI likely know they can’t keep.
Silicon Valley will continue to suffer as long as we entrust the future to management consultants and showmen who don’t build software.
Just look at Humane, a company that raised hundreds of millions of dollars to make a voice-activated AI-powered pin that ultra-popular YouTuber Marques Brownlee called “the worst product he’d ever reviewed.” One might wonder how a company would willingly launch a $700 product that overheats within minutes of use and repeatedly fails to answer basic queries, and the answer’s really simple. Humane was founded by a Bethany Bongiorno, a former management consultant for PwC and Imran Chaudhri, a former “Director of Design” from Apple that refers to himself as an “inventor” and “innovator” that was fired in 2017 for sending out an email about his planned exit from the company that suggested that Apple could no longer innovate.
When you’re insulated from real people’s problems and don’t participate in the process that might actually solve them, you’re fundamentally disconnected from any real value creation. Silicon Valley is atrophying as a result of lazy, disconnected venture capitalists and power players elevating men like Sam Altman to positions of power, and incumbents helping career consultants dictate the actions of those who actually build software and hardware.
If the tech industry wants to escape the public’s ire, it should push back against managerial poison, and talk to real people with real problems and focus on solving those before creating yet another growth-centric investor-friendly doodad.
And if the Valley really wants to change, it needs to stop empowering those who have failed upwards just because they say the right things.
