Everything in society can seem geared towards optimization, whether it’s standardized testing or artificial intelligence algorithms. We learn to know what end result you need to achieve and find a path to attain that goal.
Kenneth Stanleyformer OpenAI researcher and co-founder of a recent social media platform called Maven, has been preaching for years that this manner of pondering is counterproductive, if not downright harmful. Stanley argues that as an alternative of prioritizing goals, we must always prioritize probability.
“Sometimes, to find a springboard that will lead to the things we care about, we have to get off the path of the goal and onto the path of what is interesting,” Stanley told TechCrunch in a video interview. “Serendipity is the opposite of finding something through goals.”
The idea of searching for novelty for its own sake was born as an algorithmic concept, which was called in Stanley’s research opennessa subfield of artificial intelligence research concerned with systems that “just keep producing interesting things.”
“Open systems are like artificially creative systems,” Stanley said, noting that humans, evolution and civilization are also open systems that proceed to build on each other in unexpected ways.
This algorithmic insight evolved into Stanley’s philosophy of life. In 2015, he even wrote a book about it with his former graduate student Joel Lehman, titled The Concept Takes Off, making Stanley something of an international point of interest for the cheeky concept that, in fact, you can do things just because they’re interesting, not because they’re interesting. it’s essential to achieve a specific goal.
However, in 2022, while leading the openness team at OpenAI, Stanley found himself “seething with dissatisfaction” and “had an epiphany” in which he decided to stop talking about making openness available to the wider public and as an alternative start doing something about it.
What if, he asked himself, he created a “serendipity network,” a system designed to extend the likelihood of accidents occurring so that other people could enjoy them?
So he quit his job and began creating Maven, a social network built around an open artificial intelligence algorithm that evolves in search of recent things. When they enroll, users select a series of topics they wish to follow – from neuroscience to parenting – and the algorithm shows them posts that match their interests. Today’s social media algorithms also show you stuff you is perhaps interested in, but the difference is that they are optimized to maximise user engagement, often by boosting sensational content to drive more ad views and revenue. Maven, on the other hand, doesn’t just display the hottest posts on topics you discover interesting. The algorithm displays posts based on the likelihood that you’ll discover them attractive.
Perhaps the most revolutionary solution is Maven, which abandons the current social media setup – there are no likes, upvotes, retweets or follows, and no method to get your content to the masses.
Instead, when a user posts something, the algorithm mechanically reads the content and tags it with the appropriate interests to look on these pages. Users can move the randomness slider to maneuver beyond their specific interests, and the algorithm running the platform connects users with related interests. So if you are following conversations about land use planning, for example, Maven may additionally suggest conversations about public transportation.
And while there isn’t any method to follow people on the platform, you can see and connect with other individuals who follow topics that interest you.
In many ways, Maven looks as if an antidote to today’s social media, where the “objective paradox is on full display” as people fall over themselves to create sensational content that may attract more attention and popularity.
“Echo chambers, toxicity, narcissism reinforcement and personal branding have gotten completely out of control, causing people to lose their soul and turn into brands,” Stanley said.
The addictive properties social media, harm to mental health in adolescents and adults and opportunities polarize nations is well documented. These, Stanley argues, are the unintended consequences of ambitious goals, the result of making popularity a determinant of quality.
“And then you get all this other stuff, because once you get popular, perverse incentives come in,” he said.
Stanley noted that Maven users can flag inappropriate content or misinformation when it appears, and the AI actively monitors for highly inflammatory, offensive “or worse” content. He said Maven can’t fix the horrors of human nature, but Stanley hopes that by eliminating incentives to share such content, it can change “the overall dynamics of human behavior.”
Some social media firms have tried to combat such incentives in the past. In 2019, Instagram tested hiding likes to scale back comparisons and hurt feelings about attaching popularity to content. X, formerly Twitter, is preparing to make likes private too, but for less healthful reasons. In a line of pondering very much inspired by Elon Musk, the goal of X is to extend engagement by allowing people to privately like “edgy” content that they might otherwise not wish to protect their public image.
Maven is less interested in connecting users with audiences and more focused on connecting them with what’s interesting.
Monetization problem
Stanley and his co-founders – Blas Moros and Jimmy Secretan – launched the soft version of Maven at the end of January. The platform made its public debut in May along with, among others, Wire a feature that Stanley says has catapulted Maven to the top of the trend chart Product hunting and brought in hundreds of registrations.
These are still small numbers in comparison with other recent entrants in the social media space. The Bluesky service, which launched in 2021, has collected 5.6 million registrations. From January 2024 Mastodon had 1.8 million energetic users. Farcaster, a recent crypto-based social protocol that just raised $150 million, has counted about 350,000 signups. All these recent networks might want to expand significantly if they are to be considered effective.
It’s still an open query whether Maven will have the option to grow its user base at all without the very toxic traits that we like to hate, but which nonetheless drag us back into the cesspool that is social media.
Stanley told TechCrunch that Maven raised $2 million in 2023 in a round led by Twitter co-founder Ev Williams. OpenAI CEO Sam Altman also participated in the round. Stanley said Williams and Altman invested because, like many of us who have been captivated by Maven’s almost too-cute-for-this-world ethos, they imagine the world and the Internet need something like this.
And indeed, Maven’s idealistic hope of attracting people to interesting ideas is a breath of fresh air that smells of the early twenty first century, when the Internet was a place of contact and exploration. The sentiment of the platform’s early adopters is largely positive and upbeat, as many have flocked to the platform for its authentic and unexpected interactions and its promised freedom from toxicity.
But will the idealism be enough to draw more institutional investors later when Maven desires to expand?
“I think the challenge we face is that in the future, how to raise money will become more and more difficult,” Stanley said, noting that investors won’t throw away thousands and thousands if there isn’t any clear path to return on investment.
“I just need to find the right investors and quickly move to a sustainable business model,” he continued, musing on the idea of a subscription model that may allow Maven to maintain its ideology intact.
There are, of course, other ways for Maven to generate revenue. Advertising is one avenue, but it appeals to Stanley less because of its association with virality and sensationalism.
In the future, Maven could also potentially sell its data to firms like OpenAI that train their algorithms on data files. Earlier this month, OpenAI signed a deal with Reddit to coach artificial intelligence on the company’s social media data. And the value of Maven from the point of view of artificial intelligence is not only the content of the platform – it is the open algorithm running on it.
Stanley told TechCrunch that he believes openness is essential for artificial general intelligence (AGI), a type of artificial intelligence that goals to match or exceed human capabilities on a range of cognitive tasks. Openness is “an essential aspect of intelligence,” Stanley said. “It’s such a creative yet curiosity-driven aspect of being human.”
“Data is interesting from an AI perspective because it is about what is interesting,” Stanley said, noting that current AI models lack an intuitive understanding of what is interesting and what is not, and how which may change in time. However, while the data has potential value for artificial intelligence, Stanley said Maven does not have a contract with any company that may allow access to the data.
And although he stated that he didn’t rule out such a possibility in the future, he would consider very rigorously what the consequences of sharing such data can be.
“For me, that’s not the point,” he said, noting that he wasn’t convinced it might be a good idea for neural networks to be completely open, as that might render all human creative efforts completely pointless.
“I really wanted to create this worldwide, unexpected community,” he said. “It’s not like I have a side plan that we’re going to use Maven to create open AI or anything like that. I just wanted to create something for people because I started to feel like everyone was going to talk to chatbots more and more and we were going to be less and less connected to other people. I contributed to this as an artificial intelligence researcher.”
“Something about this idea of a web of serendipity made me feel morally better, like I could actually help make people more connected, not less.”