Cross-posted from LessWrong.
This is a description of a project I’ve been thinking about making in the last few weeks. I’m looking for feedback and collaborators. [BetterDiscourse] name is provisional.
Social media promote tribalism and polarisation. Communities drift towards groupthink even if they were specifically founded to be strongholds for rational discourse and good epistemics.
Comments sections to media articles and YouTube videos are frequently dominated by support of the author (or the pundit guest of the show), finding rational critique there is time consuming, like finding a needle in a haystack.
X’s Community Notes, pol.is, and its development variant viewpoints.xyz are universally louded, but they all use the principle of “finding the most uncontroversial common ground” which is by definition low-information for the discourse participants (because most of them should already be expected to be on board with this common ground).
Big Audacious Goal aka mission statement: improve the global online discourse. Among humans, anyway, but perhaps AI debate could benefit from the mechanics that I describe below as well.
Each user has a state-space model (SSMs are on fire right now) that represents their levels of knowledge in this or that fields, beliefs, current interests, ethics, and aesthetics, in order to predict whether the user will find a particular comment insightful/interesting/surprising, clarifying, mind-changing, or reconciling.
To provide feedback to the model, a browser extension adds the corresponding reactions (along with negative counterparts to the above reactions: “Nothing new”, “Unclear/Muddled”, “Disagree”, “Combative/Inflaming”) to the comment on the popular platforms: Reddit, YouTube, X/Twitter. (During a later stage, [BetterDiscourse] may also host comments for media which don’t have comments themselves, such as NYT or YouTube videos with comments disabled, and display them through the same browser extension, like glasp.co is doing.)
Then, when the user opens a media or a comment section on some site, the browser extension simply sorts the comments for them.
I think this should very valuable already in this completely local regime, however, things may get even more interesting, and to recapitulate the “collaborative filtration power” of Community Notes, Pol.is, and Viewpoints.xyz, (active) users’ feedbacks are aggregated to bubble up the best comments up for new users, or for users who choose not to vote actively to tune their predictive model well. Furthermore, when users with a similar state-space already voted positively for comments that their models didn’t predict then such comments could be shown earlier to other users in the same state-space cluster, overriding the predictions of their models.
More concretely, [BetterDiscourse] creates two models:
- The predictive model inputs user’s state, language embedding of the comment (or the original piece of content itself, such as an article), native rating/vote signals for the comment on the host platform, and existing reactions to the comment on [BetterDiscourse]; outputs expected user’s reactions to the content or comment.
- The transition model inputs user’s state, language embedding of the content, and user’s actual reactions to the content (or lack thereof), and outputs the next user’s state.
Data and delivery
The user’s state and model inference should, of course, stay local to the user, and the models themselves open sourced.
The initial data for training the models should either be crowdsourced (harder and longer), or collected with the assistance of LessWrong (easier). As you may have noticed, LessWrong already has all the needed reactions. Actually, I’d say that LessWrong has too many of them, I think collapsing them to just 8 mentioned above, plus general upvote and downvote would be enough.
LessWrong is also uniquely well suited for the bootstrapping data for training the models because I think due to the inclinations of the audience, just putting all these reactions on the constant display in the UI (along up/down and agree/disagree that are already there, it’s just three more pairs) should be sufficient to incentivise the users to react to the comments with the needed density.
Another aspect that makes LessWrong an ideal beach-head community for testing the mechanics of [BetterDiscourse] is that LessWrong is a relatively small community, so testing the advantages of [BetterDiscourse] for new users or users who don’t actively vote (or simply un-logged in readers) doesn’t require too big critical mass of initial active voters.
Still, LessWrong (or crowdsourced) data would be sufficient to train only a relatively small and not very capable predictive model. To make the predictive model more powerful, the SSM should be continuously re-trained on the data from at least a large portion of production users. I didn’t think hard about this yet, and I’m not an expert in federated learning and differential privacy, but I bet there should be some way to organise the model training in a privacy-preserving (and benefit-distributing, see below) way.
Reading comments and diligent voting is time-consuming work. It should be compensated by a token, similar to Brave’s Basic Attention Token.
Reactions to comments to a particular content piece are aggregated and distributed by special types of nodes (let’s call them Story Nodes), which could perhaps itself be open sourced, and spun up for arbitrary communities, including those where the content and comments are not public.
Story Nodes charge the token for accessing the best consensus sorting of comments. It is bought by people who don’t want to spend a lot of time reading through the comments and voting, but want to access comments with the best ordering, saving their time and maximising value from reading. Story Nodes share the tokens with the users who reacted to the comments “well”.
Credit assignment to the users is not trivial here: to prevent abuse of the system with junk voting, it should be based on free energy reduction (FER), i. e., formally, just the difference in the free energy of the Story Node before and after receiving a reaction from the user.
Users who intend to make [BetterDiscourse] their income source (perhaps, there should be such users to reach a sufficient voting density on many resources) will try to react to comments that don’t have many reactions yet (to maximise Story Node’s FER from their reaction) and that are expected to still receive a lot of readership in the future: Story Nodes will pay out to the voters based on their expected token revenue projects from the moment of receiving reactions.
Story Nodes are spun up by Story Node operators when they predict the story is going to be popular (e.g., a new video on a popular YouTube channel). When the attention to the story has died out, the Story Node is shut down, thus there are no amortised storage costs.
The real question is whether the cost of all inference happening on all nodes involved in [BetterDiscourse] has dropped below the market value that [BetterDiscourse] delivers which is optimising time and value from reading comments. I’m not sure of that. But even if this is currently (end of 2023) not the case, we should expect this to become so in one or two years, as the cost of inference continues to drop exponentially.
Secondary data uses and cross-sector integrations
The predictive model could directly be used for prioritising content the user doesn’t have time to process in full, such as newsletters, posts on Telegram, Mastodon and RSS feeds. This is not the initial use-case because this would probably require very big and powerful SSM to work well. Such a big SSM couldn’t be trained initially. Also, the impact is limited because the closed platforms such as Facebook, X/Twitter, YouTube, and Reddit won’t permit accessing their feeds programmatically.
Building the whole content recommendation and delivery platform (such as Substack) should be treated as a separate problem. If [BetterDiscourse] gains a lot of users, these platforms will themselves be interested to buy reactions from some [BetterDiscourse] to train their recommendation models such that if the platform later onboards a person who is already a [BetterDiscourse] user, they will receive good recommendations from the very beginning, and then continuously respond to their changing state.
Apart from Story Nodes and content platforms, users could also sell their reactions to market researchers and political researchers, or donate it to social scientists and alignment researchers.
There are also many opportunities for leveraging user’s state as inputs for other personal AI apps, such as networking (dialogue-matching) app, movie or event recommendation app, psychotherapy app, or a general personal assistant.
Spam and manipulation
There are a few strategies to combat spam and opinion manipulation on [BetterDiscourse].
However, this is not necessary in principle, and I’m curious what will happen with the discourse if AI agents can participate as voters, apart from depriving real human users of all opportunity to earn by using the system. We can speculate how this could lead to the emergence of “AI opinion leaders”, or be connected with debate, but this is outside of the scope of this post.
If voting is human-only, people could try to game [BetterDiscourse] for earning more tokens. If that happens to be the case, some standard anti-fraud data analysis can be used, and financed by Story Node operators.
The main risk that I see in the whole project is the product risk: that is, people won’t want to prioritise informationally best comments, and that their main motivation for reading comments is confirming their pre-existing worldviews. This is sort of what is customary to expect, but leaning into my optimism bias, I should plan as if this is not the case. (Otherwise, aren’t we all doomed, anyway?)
If it proves that letting AI to react to comments leads to bad results or rampant spam and manipulation (battling which requires so much compute that it makes the whole system uneconomical), gatekeeping reactions aggregation on proof-of-humanness will limit the initial reach of the system a lot, i.e., only to the people who are already have a proof-of-humanness ID, which is well less than 1% of people today.
If you are interested in building [BetterDiscourse], please leave a comment, or send me an e-mail at email@example.com.
YouTube makes this harder by not loading many comments at once.
Even this is not the end of the credit assignment story. For example, the credit that initially went to a good-sounding but misleading comment should be forwarded to “rebuttal” replies. I don’t have a complete vision of credit assignment in [BetterDiscourse] yet.