On November 15, Peter Wang posted a message requesting ideas for a new incubator and fund to support experimental projects built on the burgeoning Bluesky/AT Protocol ecosystem. Four weeks later, Skyseed emerged with an initial commitment of $1 million.
This turnaround, a speed underscored by the fact that the fund doesn’t even have a website yet (it does have a Bluesky profile, though), is testament to several things: the hype around Bluesky, which is emerging as a lifeboat for millions who have abandoned X (née Twitter). But there’s also an almost tangible hope that by starting afresh on a new social platform built on an open, decentralized network such as the AT Protocol, we might avoid the ad-driven, walled gardens that permeate social networking today.
“The vast majority of Facebook’s revenue is from ads. All the major centralized social media companies are advertising companies, which means they trade in traffic and the users’ attention,” Wang told TechCrunch in an interview this week. “The main difference between an open protocol versus a closed one is that the closed ones will never tolerate the existence of applications that take the content of the graph away, and take user-attention away from their properties.”
The decentralized web
Wang is co-founder, chief AI and innovation officer, and former CEO of Anaconda, a company built upon the eponymous open source Python and R distribution that helps data scientists build, test, and deploy all their data-driven projects.
Separately, Wang has been an avid backer of the decentralized web, providing financial support for projects such as Blue Link Labs, which developed the peer-to-peer open source web browser Beaker. Official support for Beaker ended in 2022, with creator Paul Frazee joining Bluesky as “protocol engineer” before formally becoming CTO in April.
And this foundational work on Beaker was to Bluesky’s benefit. “While the Beaker project is coming to an official end, the heart of Beaker continues with Bluesky,” Frazee wrote in a post announcing the end of Beaker. “I hope the work we do will make Beaker’s end a little less painful in the long run.”
Fast-forward to today, and Wang is extending his support for decentralization into a formal seed fund. “I’ve done angel investing, but this is my first time creating a fund and taking responsibility for other people’s money,” Wang said. “I was funding Paul [Frazee] and his team for a number of years. They tried different things to build decentralized web technologies, with middling to low success. But all those lessons learned, I think, have fed into the [Bluesky] design, the protocol, and the apps.”
While the decentralized web is far from a novel concept, what it has lacked is a meaningful number of people gravitating toward something like the AT Protocol: an open source, open standards-based framework that promises the ability to enable users to retain ownership of their data and shift to other platforms on the protocol.
“There are people who have been in the decentralized web community for a long time, who have really waited for this moment,” Wang said. “We’ve had lots of good ideas; we’ve just not had the users. Now we have the users.”
“Like the internet in 1996”
After posting his request for ideas in mid-November, Wang says he was inundated with people reaching out to him with proposals, noting that the enthusiasm reminded him of the internet in 1996.
“With all these technical ecosystems, you really need those early adopters and innovators to feel empowered to go and do all the creative things they want to do,” Wang said. “When I started seeing the quality of some of these early things that were being built, it became clear to me that something very good could come out of this.”
Digging into the makeup of Skyseed reveals two core components: the fund and the incubator. The majority of the initial $1 million comes from Wang himself, though he says some angel friends have also put in six-figure sums. Moreover, he says the fund has actually grown closer to $1.5 million, a figure that could stretch higher if current momentum maintains, with potential for formal accredited investors to get involved.
“It is mostly angels for now, but as soon as I made the announcement, I had maybe half a dozen people express interest in the fund as limited partners,” Wang said. “Just seeing the traction now, I’m thinking there are a lot more ways to deploy capital, and there’s a lot more capital that’s interested in this early-stage stuff.”
Initial checks are likely to be in the region of $100,000, though this will be directed at projects that have “real business models, real teams, and real products,” Wang said, enabling them to get to a certain proof point and raise more serious follow-on funding.
But how big can Skyseed itself get as a fund?
“I think we could get to $5 [million] to $10 million, especially as we start seeing some proof points here,” Wang said.
In addition to making formal equity investments, Skyseed will also be issuing developer grants (like Bluesky itself is doing), which will be in the region of $5,000 to $25,000.
“The grants are quite selective, in the sense that I am just turning over money to someone. I have to really trust their technical vision and ability to execute,” Wang said. “So generally, it will be people who already have some tools they’ve put out there that I think have quality and merit — either I’m using them, or I know people who are using them.”
Some developer grants will be doled out by the end of the year, with the main equity capital likely to begin trickling out early in the new year.
“I’ve already got my eyes on a few projects that are obviously great. They’re not gonna be too capital intensive, and if I can get someone off of ‘ramen mode’ just to build and focus, then that’s fine,” Wang added. “A lot of these projects don’t need a lot of funding to get far, because they’re just one person, or maybe a small two-person team working in their spare time.”
On the incubator side, meanwhile, Wang says that Skyseed will operate as a mechanism for like-minded people to bounce ideas off each other, with a view toward collaboration — particularly where complementary projects exist.
“It will be a fairly active mentorship, and it’ll be a network where they’re helping each other,” Wang said. “What I’m seeing is that there are a lot of projects in different spaces, which have similar things they want to do. So some of this is also going to be founder matchmaking.”
Dunbar’s number
Bluesky, at the end of the day, is just another app, and it’s already facing some of the same moderation challenges that X has faced almost since its inception, and advertising might land on the platform, too. But this highlights some of the potential benefits of a proper ecosystem cropping up off the back of the AT Protocol.
Dunbar’s number is a concept from British biological anthropologist Robin Dunbar that suggests a cognitive limit to how many humans can stably exist within a given social context. That number is 150. Social media plays fast-and-loose with that figure, propelling billions of people into the same social sphere, with chaos in hot pursuit; large-scale social networks force people to engage with people who they would likely avoid in real life. The AT Protocol upends that, allowing all manner of niche and micro-networks built on the same underlying framework.
“We don’t have to all fit inside the same, one-size-fits-all app,” Wang said. “The whole point of the protocol is to build many different kinds of apps, and let a lot of different kinds of user experiences flourish.”
In the few days following the launch, Wang says he had in the region of 50 projects come in, and though he was unable to confirm what projects will ultimately get funded, he did share some of his ideas on the kinds of tools that might emerge if the AT Protocol is given the space to breathe.
Alternatives to Bluesky itself are definitely on the agenda as far as the AT Protocol is concerned. This could be more of a child- and family-focused incarnation, with kid social graphs branching off of parents’ social graphs and granular controls over data privacy, for example. Or it could be something focused on dissident journalism or whistleblowers, or maybe something for marginalized communities.
Equally, it could be something akin to a Reddit clone, or socially infused alternatives to the likes of SoundCloud, Bandcamp, or even Google Maps. “Some folks have built photo-centric apps similar to Instagram, others are looking to build event-coordination platforms like Eventbrite or Partiful that are social graph-infused,” Wang said.
While it remains to be seen how such products might actually look, they could also serve as the infrastructure for an entirely new ecosystem of apps to grow on. And this gets to the heart of what Wang is striving for with his fund.
“So all these things that currently exist in the beating heart of all the big social media companies might now invert out, and become pieces of an ecosystem that is modular and which other people can build on,” Wang said.
Importantly, this doesn’t rule out ad-supported networks or subscriptions — it just gives options.
“You absolutely can have advertising-supported things, but you have to do it in balance as a business model,” Wang said. “The unfortunate thing about the big social media companies is that they stumbled into advertising and the attention economy, and by that time it was too late, they already owed billions of dollars to investors. So that’s the primary difference. We’re just going to try a different evolutionary path.”
Forked off
Despite all the brouhaha about openness and decentralization, there’s no ignoring the elephant in the room: Bluesky, controllers of not only its own app but also the underlying open source protocol, is a privately held company with the usual array of VC backers. And as we’ve seen countless times through the years, open source isn’t always permanent.
Put simply, there is plenty of scope for things to go sideways, both at Bluesky and the protocol on which it’s built. But Wang doesn’t see this as a problem.
“In almost all open source ecosystems, users are the final jury,” he said. “If a big company decides to strip-mine an open source project, if none of the users use it, then it doesn’t matter. And if users rebel against it, then they can make a fork and use that. You always have this right to exit and do alternative innovation.”
While a lot of the projects that might get funded will be spare-time side hussles, the same could also be said about Skyseed itself. It all sounds like a hell of a lot of work, with no managing partners or admins currently on the books.
“It really is something I’m doing on the side,” Wang said. “But there are a lot of allies, and a lot of people who are hopeful that this will go well. I’ve been unapologetic about grabbing their time. Part of what I’m going to be doing here over the holidays is getting some of that structure set up.”
