It would be a great way to actually learn things if, rather than deleting it, it rejected it. So, you could formulate lots of assertions, toss them at the machine and it would tell you which ones are crazy and which ones might be closer to consensus truth. You might start with all politicians are lizard people. [Bzzzzt] OK. All politicians are Jews. [Bzzzt] Hm. All politicians are liars. [Ding] Pluto is a planet. [Bzzzt] Pluto is an asteroid. [Bzzzt] Pluto is a Roman god. [Ding] Roman gods are planets. [Bzzzt] Roman gods are asteroids. [Bzzzt] Sigh. We need to parallelize this thing.
On 11/5/20 12:40 PM, Marcus Daniels wrote: > I like to imagine an open source platform that Facebook would be required to > use that would detect and label (or delete) crazy things. Like a spelling > corrector but for non-facts and confused thinking. Heavy on automated > natural language processing, but with human oversight and an audit trail. -- ↙↙↙ uǝlƃ - .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. . FRIAM Applied Complexity Group listserv Zoom Fridays 9:30a-12p Mtn GMT-6 bit.ly/virtualfriam un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com archives: http://friam.471366.n2.nabble.com/ FRIAM-COMIC http://friam-comic.blogspot.com/
