Everything in society can seem geared toward optimization—whether it’s standardized testing or artificial intelligence algorithms. We are trained to know what result you want to achieve and to find the way to achieve it.
Kenneth Stanley, a former OpenAI researcher and co-founder of a new social media platform called Maven, has been preaching for years that this way of thinking is counterproductive, if not downright harmful. Instead of prioritizing goals, Stanley says we should prioritize serendipity.
“Sometimes to find those stepping stones that will lead to the things we care about, we have to get off the path of the goal and onto the path of the interesting,” Stanley told TechCrunch in a video interview. “Story is the opposite of finding something through goals.”
The idea of seeking novelty for its own sake began as an algorithmic concept Stanley studied called openness, a subfield of AI research into systems that “just keep producing interesting things forever.”
“Open-ended systems are like artificially creative systems,” Stanley said, noting that humans, evolution and civilization are also open-ended systems that keep building on themselves in unexpected ways.
This algorithmic insight became a life philosophy for Stanley. He even wrote a book about it in 2015 with his former PhD student, Joel Lehmann Why greatness can’t be planned. The concept took off, making Stanley something of an international focus for the cheeky idea that you can actually just do things because they’re interesting, not because you have to fulfill some stated goal.
But in 2022, while leading an open-end team at OpenAI, Stanley said he was “seething with discontent” and “had this epiphany” when he decided to stop talking about bringing open-end to a wider audience and instead to start doing something about it.
What if, he asked, he created an “intuition network,” a system that was tuned to increase the likelihood of intuition for other people to enjoy?
So he quit his job and set out to create Maven, a social network built around an open AI algorithm that evolves to search for novelties. When they sign up, users choose a series of topics to follow—from neuroscience to parenting—and the algorithm shows them posts that match their interests. There are no likes, upvotes, retweets or follows, and no way to push content to the masses.
Instead, when a user posts something, the algorithm automatically reads the content and tags it with relevant interests so that it appears on those pages. Users can increase the serendipity slider to branch out beyond their stated interests, and the algorithm driving the platform matches users with related interests. So if you’re following conversations about urban planning, for example, Maven can also suggest conversations about public transportation.
And while there’s no way to follow people on the platform, you can see and connect with other people who follow topics that interest you.
In many ways, Maven feels like an antidote to today’s social media, where “the objective paradox is on full display” as people exalt themselves to create sensational content that will attract more attention and popularity.
“The echo chambers and the toxicity and the amplification of the narcissism and the personal brand have gone completely out of control, so people are losing their souls and becoming brands,” Stanley said.
The addictive qualities of social media, the damage to the mental health of adolescents and adults, and the ability to polarize nations are well documented. These, Stanley says, are the unintended consequences of ambitious goals, the result of making popularity a substitute for quality.
“And then you get all these other things, because once you have popularity, you have spin-off incentives,” he said.
Stanley noted that Maven users can flag inappropriate content or misinformation as it occurs, and its AI actively monitors for highly inflammatory, offensive “or worse” content. He said Maven can’t fix what’s wrong with human nature, but by removing the incentives behind sharing such content, Stanley hopes it can change “the overall aggregate dynamics of how people behave.”
Some social media companies have tried to combat such incentives in the past. The OG for pushing out random content was StumbleUpon, a browser extension and app created by entrepreneur Garrett Camp years before he co-founded Uber. Then in 2019, Instagram tested hiding “likes” to limit the comparisons and hurt feelings that come with making content popular. X, formerly Twitter, is also preparing to make likes private, but for less salutary reasons. In a very Elon Musk-inspired line of thinking, X’s goal is to create greater engagement by allowing people to personally like “edgy” content they otherwise wouldn’t in order to protect their public image.
Maven is less interested in connecting users with audiences and more focused on connecting them with what’s interesting.
The problem with monetization
Stanley and his co-founders – Blas Moros and Jimmy Secreton – soft-launched Maven in late January. The platform debuted publicly in May alongside a Wired feature that Stanley says gave Maven the most popular spot on Product Hunt and led to thousands of signups.
These are still small numbers compared to other new entrants into the social media space. Bluesky, which launched in 2021, has 5.6 million registrations. As of January 2024, Mastodon has 1.8 million active users. Farcaster, a new crypto-based social protocol that just raised $150 million, has counted around 350,000 registrations. All of these new networks will need to grow significantly if they are to be considered successful.
It’s still an open question whether Maven will even be able to grow its user base without the highly toxic qualities that we love to hate, but which nevertheless drag us back to the cesspool that is social media.
Maven raised $2 million in 2023 in a round led by Twitter co-founder Ev Williams, Stanley told TechCrunch. OpenAI CEO Sam Altman also participated in the round. Stanley said Williams and Altman invested because, like many of us who have become enamored with Maven’s almost-too-sweet-for-this-world spirit, they felt the world and the Internet needed something similar.
And indeed, Maven’s idealistic hope to connect people with interesting ideas is reminiscent of the early 2000s, when the Internet was a place to connect and explore. The sentiment of early users of the platform is mostly positive and optimistic, as many of them came to the platform for genuine and casual interactions and the promised freedom from toxicity.
But will the idealism be enough to attract more institutional investors later when Maven wants to grow?
“I think the challenge we face is that going forward it’s going to be an increasingly difficult way to raise money,” Stanley said, noting that investors won’t throw in millions unless there’s a clear path to getting a return on your investments.
“I just need to find the right investors going forward and quickly get to a sustainable business model,” he continued, considering the idea of a subscription model that would allow Maven to keep its ideology intact.
There are, of course, other ways for Maven to bring in revenue. Advertising is one avenue, but one that is less appealing to Stanley because of how much it is tied to virality and sensationalism.
Ultimately, Maven could potentially sell its data to companies like OpenAI, which train their algorithms on reams of data. OpenAI earlier this month signed a deal with Reddit to train its AI on the social media company’s data. And Maven’s value proposition from an AI perspective isn’t even just the content of the platform—it’s the open-ended algorithm that runs it.
Stanley told TechCrunch that he believes the open end is essential for artificial general intelligence (AGI), a type of AI that aims to match or surpass human abilities in a range of cognitive tasks. Openness is “such an important aspect of intelligence,” Stanley said. “It’s like that creative and curiosity-driven aspect of being human.”
“Data is interesting from an AI perspective because it’s data about what’s interesting,” Stanley said, noting that current AI models lack an intuitive understanding of what’s interesting and what’s not, and how that can be changes over time. However, while the data has potential value for AI, Stanley said Maven does not have a deal with any company to provide access to this data.
And while he said he hasn’t ruled out the possibility in the future, he will think very carefully about what the implications of sharing such data would be.
“That’s not the point of it for me,” he said, noting that he’s not convinced that it would be a good idea to make neural networks completely open, because that could render any creative human endeavors completely meaningless.
“I really wanted to create this worldwide casual community,” he said. “It’s not like I have a side plan that we’re going to use Maven to create open-ended AI or anything like that. I just wanted to create something for people because I started to feel that everyone would talk to chatbots more and more and we would be less and less connected to other people. And I was contributing to that as an artificial intelligence researcher.”
“Something about this idea of a web of randomness made me feel morally better, like I could actually contribute to people being more connected, not less.”
