Everything in society can be oriented towards optimization – whether it’s standardized tests or artificial intelligence algorithms. We are taught to think about what outcome you want to achieve, and find a way to get there.
Kenneth StanleyA former OpenAI researcher and co-founder of a new social media platform MavenThis way of thinking is counterproductive, if not outright harmful, he has been preaching for years. Instead of prioritizing motives, Stanley says we should prioritize coincidence.
“Sometimes, to find the steps that will lead us to the things we care about, we have to get off the path of purpose and take the interesting path,” Stanley told TechCrunch in a video interview. “Happiness is the opposite of getting something through purpose.”
The idea of seeking novelty in itself began as an algorithmic concept that the Stanley study called open-endednessA subfield of AI research about systems that are “always producing interesting things.”
Stanley said, “Open-ended systems are like artificially creative systems”, and explained that humans, evolution, and civilization are all open-ended systems that continue to create themselves in unpredictable ways.
This algorithmic insight turned into a life philosophy for Stanley. He even wrote a book on it in 2015 with his former PhD student Joel Lehman called Why can’t greatness be planned? The concept caught on, and Stanley became an international focal point for the brazen idea that, in fact, you can do things simply because they’re interesting, not because you have to accomplish some stated purpose.
But while leading the open-endedness team at OpenAI in 2022, Stanley said he was “boiling with dissatisfaction” and “came to the realization” that he stopped talking about bringing open-endedness to a wider audience and started doing something about it instead.
He asked himself, what if he created a “serendipity network,” a system set up to increase the likelihood of serendipity so that other people could enjoy it?
So he quit his job and set out to build a social network called Maven, built around an open-ended AI algorithm that evolves in search of novelty. When signing up, users select a range of topics to follow — from neuroscience to parenting — and the algorithm shows them posts that match their interests. Today’s social media algorithms also show you things you might find interesting, but the difference is that they’re optimized to maximize user engagement, often by promoting sensational content, to make more ad impressions and revenue. In contrast, Maven doesn’t just show you the most popular posts on topics you find interesting. The algorithm shows you posts based on the likelihood that you’ll find them engaging.
Perhaps most revolutionary of all, Maven has done away with the current system of social media – there are no likes, upvotes, retweets or follows, and no way to distribute content to the general public.
Instead, when a user posts something, the algorithm automatically reads the content and tags it with relevant interests so it shows up on those pages. Users can turn on the Serendipity slider to go beyond their stated interests, and the algorithm running the platform connects users with related interests. So, for example, if you’re following a conversation about urban planning, Maven might also suggest a conversation about public transportation.
And while there’s no way to follow people on this platform, you can still see and connect with people who follow topics that interest you.
In many ways, Maven sounds like an antidote to today’s social media, where “the objective paradox is on full display” as people push themselves to create sensational content that will gain more attention and popularity.
“Echo chambers and toxicity, the amplification of narcissism, and personal branding have gotten completely out of control, causing people to lose their souls and turn into brands,” Stanley said.
Addictive properties of social media, Harm to mental health in adolescents and adults, and the ability to Polarization of nations These, Stanley says, are the unintended consequences of ambitious goals, the result of making popularity a proxy for quality.
“And then you get all these other things, because once you have popularity, you also get perverse incentives,” he added.
Stanley said Maven’s users can flag inappropriate content or misinformation when it comes across, and its AI actively monitors for content that is excessively inflammatory, offensive “or worse.” Maven can’t fix the wickedness that exists in human nature, he said, but by eliminating the incentive behind sharing such content, Stanley hopes it can change “the overall overall dynamics of people’s behavior.”
Some social media companies have attempted to tackle such incentives in the past. Instagram in 2019 Hiding likes was tested to prevent comparisons and hurt feelings caused by associating content with popularity. X, formerly known as Twitter, Preparations to make likes privatetoo, but for less healthy reasons. Inspired by Elon Musk, X aims to generate more engagement by allowing people to privately “edgy” content they might not otherwise like in order to protect their public image.
Maven is less interested in connecting users to an audience, and more focused on connecting them to things that are interesting.
Monetization problem
Stanley and his co-founders — Blas Moros and Jimmy Secretan — soft-launched Maven in late January. The platform was publicly launched in May Wired Stanley says this feature catapulted Maven to the top trending spot Product Hunt And thousands of people registered.
These numbers are still small compared to other new entrants in the social media space. Blue SkyLaunched in 2021, this app has 5.6 million sign ups. By January 2024, Mastodon had 1.8 million customers Active users. Farcaster, a new crypto-based social protocol that Just raised $150 millionhas counted around 350,000 signups. All of these new networks will need to grow significantly to be considered successful.
It’s still an open question whether Maven will be able to grow its user base without the toxic qualities we hate, but which lead us back into the filth pit that is social media.
Maven, led by Twitter co-founder Ev Williams, raised $2 million in a 2023 round, Stanley told TechCrunch. OpenAI CEO Sam Altman also participated in the round. Stanley said Williams and Altman invested because while many of us have been impressed by Maven’s almost too sweet for the world ethos, they think the world and the internet need something like this.
And indeed, Maven’s idealistic hope of connecting people with interesting ideas is a breath of fresh air that smells like the early 2000s, when the internet was a place of connection and exploration. Sentiments from early users on the platform are mostly positive and optimistic, as many came to the platform for the promise of genuine and casual conversation and freedom from toxicity.
But will that idealism be enough to bring in more institutional investors later, when Maven wants to move forward?
“I think the challenge we face is that this method of raising money is going to be even more difficult going forward,” Stanley said, adding that investors won’t put in millions of dollars unless there’s a clear path to getting a return on their investment.
“I just need to find the right investors to move forward and quickly create a sustainable business model,” he added, considering the idea of a subscription model that would allow Maven to keep its ideology intact.
Of course, there are other ways for Maven to bring in revenue. Advertising is one way, but it appeals less to Stanley because it’s so tied to virality and sensationalism.
Going forward, Maven could potentially also sell its data to companies like OpenAI that are training their algorithms on a pile 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 just limited to the content on the platform—it’s the open-ended algorithm that drives it.
Stanley told TechCrunch he believes open-endedness is crucial for artificial general intelligence (AGI), a type of AI that aims to match or surpass human abilities across a range of cognitive tasks. Open-endedness is “such a key aspect of being intelligent,” Stanley said. “It’s the creative and curiosity-driven aspect of being human.”
“This data is interesting from an AI perspective, because it’s data about interesting things,” Stanley said, adding that current AI models are losing an intuitive understanding of what’s interesting and what’s not, and how that might change over time. However, even if the data has potential value for AI, Stanley said Maven hasn’t struck a deal with any company to give it access to that data.
Although he said he had not ruled out such a possibility in the future, he would consider very carefully what the consequences of sharing such data would be.
“That’s not the point for me,” he said, adding that he doesn’t agree that having neural networks completely open would be a good thing, because that would make any creative effort by humans completely redundant.
“I really wanted to build this worldwide casual community,” he said. “It’s not like I had some side plan that we were going to use Maven to build open-ended AI or something. I just wanted to build something for people because I was starting to think that everyone was going to be talking to chatbots more and more and we’d be connecting with other people less and less. And I was contributing to that by being an AI researcher.”
“The idea of this unexpected network made me feel morally superior, like I could actually contribute to connecting people more rather than connecting them less and less.”
how-mavens-ai-run-serendipity-network-can-make-social-media-interesting-again-techcrunch
