PMC COVlD Update, Oct 26, 2026 🔥44 states with COVlD outbreaks (see Alt) 🔥Double wave peaks, both at an estimated 950,000 new daily infections 🔥6th week in a row of massive CDC upward retroactive corrections going back to early July THREAD (1/4)

Subtweet: Imagine you had a big heart and built a Discord/Reddit following trying to help people avoid COVID. BUT you oversold certain tactics (e.g., nasal sprays, certain vaccine options), and people increasingly realize this. Now, people are using AI more, and basic AI models are better incorporating medical evidence, in part because they draw upon critiques by myself and thousands of others and evaluate those critiques, and the critiques often (but not always!) win. It's nowhere near "there" yet. It still makes big mistakes by default on "statistical conclusion validity," stating what study authors claim, even when their own data do not support it. That overly perpetuates industry bias amid sketchy studies. But it's improving, and in a year will be much better. In 5 years, it might even recommend clinicians mask. So, what do you do if you're the good-hearted person who got some things wrong and is losing clout? In science, the goal is always to be up to date on the current evidence. For example, although past nasal spray studies are flawed, if a good one with a positive trial came on board, I would eat humble pie and encourage it. I was very hopeful of this when the JAMA-IM article came out; just more junk, unfortunately. If Long COVlD incidence declined, I would warn of that too. These would both be hugely positive developments. Stating one perspective and then admitting being wrong based on new data often gives one immense credibility and is extremely helpful. It's like when a politician votes against the rest of their party. It's just good science, and we're in a terrible public health place because so few people do this. So, if one oversold nasal sprays and certain vaccine options based on good faith and a big heart, and people were increasingly realizing this, the simplest response would just be to acknowledge the cumulative body of evidence. Don't oversell. Don't attack scientists. Don't push a biased lit review. That will all fail. Just acknowledge what's evolved and move on.

PMC COVlD Update, Oct 17, 2026 (U.S.) Early Release COVlD levels have rebounded into a secondary peak, as the CDC continues to retroactively correct levels upward. Outbreaks span 44 states (see Alt). Levels are falling in TX, still high out west, & up in the Northeast. 🧵1/3