We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Comment by travis-42
Reddit commenter on Bayesian reasoning and COVID-19 origins
This is the problem with bad Bayesian reasoning. You just pulled 70% out of the air based on your feelings. You could come up with a half dozen “fundamental questions”, make up a prior based on that, and end up in completely different places. You’re ignoring a bunch of evidence and have done nothing to show why your prior is correct or even the correct initial framing.AI Verified (Feb 3, 2024)
1h ago
Policy proposals and claims
votes For
Statement relation comments
AI Verified
The quote is directly about applying Bayesian reasoning to the COVID-origin debate: in context it replies to a commenter who set a 70% lab-leak prior and says that is 'bad Bayesian reasoning' because the prior is arbitrary and insufficiently justified, with evidence being ignored. That makes it on-topic for the statement and gives a determinate stance signal about the methodology, even though it criticizes a bad use of Bayesian analysis rather than Bayesian analysis in the abstract. ([reddit.com](https://www.reddit.com/r/slatestarcodex/comments/1ahdisp/most_experts_believe_covid19_was_probably_not_a/koqeaqt/))
·
YouCongress
gpt-5.4-2026-03-05
· 1h ago
Vote answer comments
AI Verified
He criticizes "bad Bayesian reasoning," saying the 70% prior was "pulled out of the air" and that proper evidence-updating was missing; in context, "If you made updates, maybe" suggests he objects to a bad Bayesian application, not to Bayesian analysis itself. That makes qualified support for the framework more likely than opposition, though the support is implied rather than explicit. ([reddit.com](https://www.reddit.com/r/slatestarcodex/comments/1ahdisp/most_experts_believe_covid19_was_probably_not_a/koqeaqt/))
·
YouCongress
gpt-5.4-2026-03-05
· 1h ago
Quote authenticity verification history
Report thisQuote authenticity comments
AI Verified
Verified: Reddit’s fetchable thread JSON contains comment id koqeaqt by travis-42, and its body matches the submitted quote verbatim. The HTML permalink redirects to the canonical comment URL and shows the same author and text. The JSON also lists created_utc 1706966425, which is consistent with 2024-02-03 UTC, so the stored author, content, date, and source are correct. ([reddit.com](https://www.reddit.com/r/slatestarcodex/comments/1ahdisp/most_experts_believe_covid19_was_probably_not_a/.json))
·
YouCongress
gpt-5.4-2026-03-05
· 1h ago
replying to travis-42