Quantifying Thought Pollution (Moloch cont.)

Quantifying Thought Pollution (Moloch cont.)

After the massive success of TikTok, other social media apps copied it at a comical speed (Youtube Shorts, Insta Reels, Snap... etc.). This sort of behavior is natural in an environment like the current social media landscape, where competitive pressures make tech giants prioritize attention above human wellbeing.

"We care deeply about issues like safety, well-being and mental health." -Mark Zuckerberg Yeah... the punchlines right themselves.

Moloch Yet Again

Competitive pressures forcing optimization at all costs is a manifestation of Moloch, by definition. Lose-lose scenarios are his domain, and social media is a place where this manifests clearly.

Media Minus Moloch?

Given the reality of competition, how could we reduce or eliminate the incentives that encourage the emergence of lose-lose scenarios?

A few ideas that spring to mind:

Attention taxes.

Similar to carbon-taxes, this approach would attempt to quantify the cost of lost human potential. 1 hour of time spent scrolling instagram is 1 hour that can't be used to learn, cure cancer, or paint murals. Similar to carbon taxes being used to offset externalized costs, attention taxes could be used to fund public goods or art projects, which would be a way of incentivizing external goods.


  • If squandered attention can be quantified well, this could be the best at harnessing the power of the free market to avoid sucking people's attention away from them.
  • Funds from attention taxes could help incentivize win-win behavior.
  • There are a few easy proxies to use for quantifying this. The best may even be the naive counting of man hours spent scrolling.


  • Quantifying distraction seems really hard. How can you distinguish between someone learning through engaging content (Youtube University), versus someone doom-scrolling?
  • Enforcing a false binary. Slack (the ability able to have time without binding constraints) is important. Not all time falls neatly into categories of "wasted" and "well-spent."

Ownership alignment

Another idea would be to have users be owners, aligning incentives. This philosophy is prominent in web3 communities.


  • Users are by definition no longer in an "other" category. Perhaps this maps to increased empathy for users?


  • Ownership doesn't solve misalignment. Public stock markets have distributed ownership of many companies. But public companies (Meta, Apple, Snap, McDonalds, etc) don't seem to treat users any better than private companies (Patagonia, Chick-fil-a, Deloitte, etc).

Bans & Access Restrictions

Many jurisdictions ban or age-restrict addictive or dangerous substances like fentanyl, alcohol, and cigarettes.

An option may be to outright ban certain features. For example, if "like" buttons are found to be psychologically destructive, maybe a good government should make them illegal or age-restricted.


  • Central coordination. Moloch rears his head due to competitive pressures between multiple actors triggering lose-lose traps. Central coordination can avoid this, ensuring that no actor is able to defect.


  • Regulatory capture. It feels very difficult to escape having any body in charge of regulating social media from piling on red tape and restrictions on innovation.
  • No clear guidelines. Jurisdictions can't even agree what non-digital items should be banned. Some jurisdictions legalize cocaine, while others punish it by death. There's no reason to suppose that the average jurisdiction will be better at determining which features should be legal.

This is a list just off the top of my head. I'm curious to see what other options are available, and any pros or cons I didn't think of. Drop me a line if you think of any interesting ones. :)

Subscribe to joshuaa.dev

Don’t miss out on the latest issues. Sign up now to get access to the library of members-only issues.