Laura Pike Seeley
@laurapikeseeley
Laura Pike Seeley
@laurapikeseeley
In her ideal world, we’d halt work on making AI more powerful for the next five to 10 years. In the meantime, society could get used to the very powerful systems we already have, and experts could do as much safety research on them as possible until they hit diminishing returns. Then they could make AI systems slightly more powerful, wait another
... See moreThe rule of reframing can look different depending on what you're dealing with. But it always involves seeing a problem from a different angle, with the goal of breaking a cycle of harmful thoughts and behavior.
Adopting a healthy scepticism towards inherited ideas means “emptying the container of the Self”, or perhaps exercising a kind of trained “forgetting” of that which is extraneous to the essential self. In this state, we become less prone to partaking in the superfluous occupations we impose upon ourselves to cope with the weight of social
... See moreExperts have already seen and documented more than 60 smaller-scale examples of AI systems trying to do something other than what their designer wants (for example, getting the high score in a video game, not by playing fairly or learning
game skills but by hacking the scoring system).