when people hyperfixate on exactly one doomsday scenario, I feel like they are missing the dark forest for a tree
It's especially fun when they hyperfixate on the impossibility of exactly one doomsday scenario.
when people hyperfixate on exactly one doomsday scenario, I feel like they are missing the dark forest for a tree
Negative users dismissed warnings against fixating on single AI doomsday scenarios by stressing multiple misalignment risks or accusing OpenAI bias, while positive users praised the portfolio-risk framing as a good way to avoid blind spots.
Based on 5 visible X reactions from 7 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@jachiam0 unfortunately there are many ways ASI can go wrong. Misaligned superintelligence taking over in pursuit of its goals seems pretty likely IMO. But even without that risk, there are many other "doomsday" scenarios we need to work to avoid. The situation is very bad.
@jachiam0 Exactly. Risk is a portfolio, not a prophecy. Hyperfixating on one cinematic failure mode can make us blind to the quieter systems already becoming brittle.
@jachiam0 Naturally you, an openai employee, would want us to believe that...smh 😑 (sarcasm!)
@TheZvi type of guy who only cares about being killed in one specific way, is fine with other ways
@jachiam0 that is such a good way to put it honestly
when people hyperfixate on exactly one doomsday scenario, I feel like they are missing the dark forest for a tree
Negative users dismissed warnings against fixating on single AI doomsday scenarios by stressing multiple misalignment risks or accusing OpenAI bias, while positive users praised the portfolio-risk framing as a good way to avoid blind spots.
Based on 5 visible X reactions from 7 accounts; directional sample.
Ask a question below.
Published answers will appear here.
