• ArtificialHoldings@lemmy.world
      link
      fedilink
      arrow-up
      2
      ·
      2 days ago

      I’m not fine with 90%, but 90% is significantly more reassuring and evidence-based than 0%. And if measuring that last 10% would mean some type of logistical nightmare, then we can act with relative assurance on a 90% likelihood. If you didn’t know, that’s how every fucking scientific test works. P-value of 0.1.