Elon is most likely buying Twitter.
Elon’s ownership will almost certainly benefit Twitter’s current and future users, especially with product improvement velocity. However, I’m ambivalent on the societal impact of this decision.
After this closes, there will only be three major social media platforms used in the western world that can be controlled by public shareholders: Pinterest, LinkedIn, and Twitch. Big in their specific categories, but not the big cultural drivers. Snapchat, Facebook, Google, and TikTok are all controlled by the supervoting shares of founders or governments. While that concentrated ownership may lower the probability of the likelihood of censorship, it certainly increases the ability.
Let’s dig into his planned improvements, including on “safety”, and what the broader societal implications could be.
Disclaimer: I’ve been an Elon and Tesla Fan for years. The soft Tesla shirt in college definitely helped! It’s because of what he seeks to achieve, his willingness to act on his conviction, and the clarity of thought. He, like everyone, isn’t perfect.
Roadmap
One of my decision making mantras when working in an area with lots of ambiguity is to “do no brainers while you get smart.”
Elon’s roadmap is filled with lots of no brainers.
Fighting spam. It’s noise, no argument here as long as you can actually do it while skirting the fine line between promotion & spam.
Temporary bans over permanent bans. Cutting people off from information and societal infrastructure is risky both for their development but also in terms of exacerbating polarization. Permanent bans have their place, but temporary is certainly where to start.
Monetize tweeting users with big followings, including governments. This answers honestly that Twitter’s advertising business is nowhere near Facebook and Google’s. The reason to be active on Twitter is the relationship and messaging with your audience that you can leverage into other revenue opportunities, similar to something like Whatsapp for Business. This revenue model is similar to an aspect of what WeChat employs.
Give Twitter Blue members a verified check mark. Payment is a pretty decent proxy for identity, especially considering Know Your Customer laws. Nice way to also create some social proof that people are buying Twitter Blue and increase subscriptions.
Enable longer tweets and longer-form consumption. Twitter threads are a cool user-originated phenomenon, like many things on Twitter, but they are sub-optimal as a consumption experience. Sometimes you read things out of sequence, especially if one part of the thread is re-tweeted.
The edit button!
“We’ve been exploring how to build an Edit feature in a safe manner since last year and plan to begin testing it within @TwitterBlue Labs in the coming months,” Twitter’s head of consumer product Jay Sullivan said in a tweet on Tuesday.
It’s still mindblowing that Twitter STILL hasn’t shipped a version of an edit button during this time. It’s been 6 weeks since he disclosed his massive stake. Twitter has supposedly been “working on it” for years. The tech team owning this feature is almost assuring their own firing by moving unfortunately slowly here. It’s a nuanced feature, especially in how you show a tweet has been edited, but there’s plenty of other products to emulate.
The not so easy
However, there’s some aspects of his plan that feel more like bluster than reality and are likely much harder than he planned.
Moderation
Social media moderation is a very difficult game of whack a mole - even if you are trying to reduce the amount of moderation. The risk vectors get to nearly infinite. You have hundreds of millions of people who can interact with hundreds of millions of other people who can cause risk/safety concerns in nearly all the nuanced ways that people can communicate.
This difficulty is furthered by the fact that expression is the core product experience. Child porn might be a very clear and high intent signal to act against. But where does discussion about reduced vaccine efficacy turn into misinformation? Shades of grey between safety and core utility.
This makes trust & safety work SUPER hard on very open-ended, unconstrained social networks. Reddit may be easier to work on, than say Facebook or Twitter, because you can change the enforced policies based on what users have opted into vis-a-vis their subreddit.
My personal suspicion is that self-driving cars are ultimately an “easier” problem than social media safety. It’s less open ended. And that’s saying something considering it’s moved years slower than originally expected. After all, Elon honestly conceded during his TED talk that the timing of self driving didn’t quite happen as he planned (see again this video starting in 2014 saying this is the year of self driving).
Elon’s stated move away from advertising could be the salve here. I’ve shared before that I do think safety, especially it’s perception, affects growth. By moving Twitter away from ad revenue, and thus have its financial security less tied to increased engagement, it can take a more open approach both to an algorithmically driven feed and the policies it enforces on its platform.
Identity
How do you “authenticate all the humans”? Especially when you’ve got a platform that has many pseudonymous accounts? It’s a good goal, especially to reduce harassment and have users to take ownership of their behavior. But feels like this isn’t an exactly easy or clear solve. Please email me back if you disagree!
Algorithmic choice
One of Elon’s plans is to open source Twitter’s ranking algorithm and put it on Github. That’s relatively clear to do, and I do think it’s possible it increases trust. But another plan is to give users choice over their algorithm. While not impossible, it’s not exactly straightforward. Users may understand ‘Most Recent’ filters, but how do you frame something like ‘Most shared by people you only occasionally engage with’ or ‘Most divisive’?
Open-ended use
Elon hasn’t built a product like Twitter before. The closest is Paypal. Cars and Rockets have a bit more objectivity and consistency of customer expectations than a hugely mass market product like Twitter. That being said, he’s clearly a power user so I think his empathy here, especially for the tweeting user, carries a lot of weight. Plus it’s not like he knew that much about rockets or cars before he started there.
Societal implications
As I said above, this concentrates ownership of another major social platform. In Twitter’s case, I think it’ll result in much better product improvements.
However, writ large, it’s a concerning trend that enables the manipulation of billions of people both through censorship as well what topics receive algorithmic amplification. It also means ownership have no accountability, other than to their own appointed bodies, about the decisions they make about who is on the platform. It’s weird that Elon can unilaterally make the decision to re-onboard a President.
If Elon can diversify Twitter away from solely advertising, it’s a massive win that enables a more unique and open-minded product experience. The ability to pick your own algorithm could be tremendously valuable for the many, many people who are seeking to round out their exposure rather than just get farther into an algorithmically amplified filter bubble.
I agree with Elon’s assertion that free, but not hateful, speech is critical to a functioning democracy. Further, commonly reviewed information, with commonly held standards, is critical to effective group decision making. That effective decision making has been unfortunately missing in US national politics. A bit less polarization could be a big unlock to make progress here.
The former CEO of Reddit, Yishan Wong, had a rambling but fascinating twitter thread about his opinions of Elon’s twitter purchase. It’s worth a read. I wish he would have gone into more specificity on why social media platforms have to moderate even when they don’t want to. Nonetheless, there was one part that really stuck out to me.
There’s something to this. Tesla is still relatively early in its journey. The climate crisis isn’t over. SpaceX hasn’t reached Mars. Can Elon divide his attention further?
One of Elon’s strengths is that when he commits to something, he usually goes after it with intensity and longevity. He’s been involved with Tesla and SpaceX for nearly 20 years and even the Boring Company has been a project for six years. We should expect no less in terms of Twitter - but I’m not totally sure that’s a good thing considering the importance of his other obligations.
Further, I worry about the impact on Tesla with so many of its shares being used as purchase collateral. That could have an impact on its cost of capital, which has been a historic strength, which in turn could affect future expansion of capabilities like factories and superchargers. While there has also been a general rout on tech equities, Tesla’s shares have fallen significantly since this plan was announced.
In closing
Overall, I’m intrigued as this could be a valuable exploration for society and probably a major improvement to the Twitter experience. It’s also fascinating we get to see this.
Elon has turned himself into a meme and an internet-enabled quasi-religious figure. He captures hearts, minds, and attention (I’m writing this after all). All of this fosters a huge cult-like, share-buying following but it also creates many haters.
He has also expanded the bounds of what people expect as possible. He did it with rocketry and space and he also did it with electric transportation. Maybe he also does it for how we publicly communicate. Even if he doesn’t, I’m willing to believe he’ll try really damn hard. I just hope it doesn’t take away from the other worthy causes where he and his companies are already winning.