Why did General Nakasone join OpenAI? Because THIS. There is a TON of work to do!
(twitter.com)
🐸 WAG THE FROG 🐸
You're viewing a single comment thread. View all comments, or full comment thread.
Comments (74)
sorted by:
There's a group of pro-vaccine professionals from various backgrounds on Twitter who I'm convinced are some form of AI.
https://twitter.com/IanCopeland5
Copeland (cope-land) in particular has stated that you'd never guess where he comes from, that he's "with the Army" and is a "PHD Level Geneticist".
If you visit their space you'll notice some strange and highly predictable patterns to their format. The rules are that the "anti-vaxx" crowd can come up to "debate" the issues, but they must only cite peer-reviewed sources from the "most-respected" scientific journals. They aren't allowed to use any logical fallacies (they are immediately muted and scolded if they do), smaller scale studies, their own reasoning, or evidence from a non-scientific context (e.g. monetary or power motivations).
Conversely--and infuriatingly--the pro-vaxx panel are able to use all of the logical fallacies--and proceed to do so judiciously. They're also permitted to cite smaller scale studies & non-RCT when it's convenient, etc.
In short, it's the most unfair debate format imaginable, and is essentially unwinnable for the anti-vaxx side because no anti-vaxx study would ever be published or funded.
But what I view it as is a white-hat created debate gauntlet. A training ground where white hats can search for new angles of attack based on the outcomes of the battles. What can humans do to give the AI a brain fart in these impossible scenarios?
They also occasionally throw in a bit of bait to make things interesting and give the combatants a bit of help (e.g. the Copeland character was given myocarditis in his 30s but vehemently denies that it's vaccine related).
You have to think there's no way so many full-time professionals would find the time to host what is essentially the same space every day to no productive end. I strongly suspect it's an AI creation.
Last thing:, the AI needs "a link" to begin, and is very insistent that participants in the debate give it a link before proceeding. Imagine an AI that is given the prompt:
^ With that you have Dr. Ian Copeland 5.
Fascinating.
very interesting
Enjoy training your replacements ?
But glad to hear you are keeping an optimistic outlook.