This week I was in a collaborative discussion about systemic fixes to healthcare.

· Bits and Bobs 9/22/25
  • This week I was in a collaborative discussion about systemic fixes to healthcare.
    • My conclusion is that it's an absurdly complex area (duh!).
    • But two ideas that stuck with me:
    • Transparency plus AI.
      • First, make risk-adjusted outcome data transparent per provider.
      • Second, allow LLMs working on behalf of a given user to research the best provider for them.
      • Humans don't have the patience to compare all of the different options, but LLMs do.
      • This would lead to competition not on cost per procedure but cost per outcome.
    • Allow employees to keep their insurance plan when they move employers.
      • Given how often people switch employers, the average length of time people have on a plan is 2-3 years.
      • That means that the insurer is structurally incentivized to under-invest in preventative care.
        • By the time the condition becomes acute, the user is likely someone else's responsibility.
      • Also, users don't really have a choice… it's whatever provider their employer picked, and that's often based on cost, not based on quality of service or outcomes.
      • But if users could keep the same plan when they shifted employers, then the average amount of time a given user stays with a plan would go up.
        • That would structurally incentivize long-term preventative care more.
        • The insurers would then have a longer-term incentive to keep the customers happy and healthy.
      • Employers would put whatever budget they would have put into an employee's health plan and pay towards whatever plan the employee picked.

More on this topic

From other episodes