The case for slowing down AI

added by Laura Pike Seeley · updated 2y ago

The problem we have in AI right now is not that it’s getting too powerful, it’s that it’s not nearly powerful enough. Very little has changed thus far because of AI, and won’t until models get faster, cheaper, more accurate, and more “intelligent”. Building safe AI is insanely important, but any goal that is half in and half out of driving progress... See more
from Tweet by Aaron Levie
sari added
But you’ll notice

what I slipped in there – I said we should focus first on preventing AI-assisted crimes before they happen – wouldn’t such prevention mean banning AI? Well, there’s another way to prevent such actions, and that’s by using AI as a defensive tool . The same capabilities that make AI dangerous in the hands of bad guys with bad goals m... See more
from Why AI Will Save the World | Andreessen Horowitz by Marc Andreessen
Parker C added
I think the most worrisome aspect of AI systems in the short term is that we will give them too much autonomy without being fully aware of their limitations and vulnerabilities. We tend to anthropomorphize AI systems: we impute human qualities to them and end up overestimating the extent to which these systems can actually be fully trusted.
from Artificial Intelligence: A Guide for Thinking Humans by Melanie Mitchell
sari and added

from Tweet by Aaron Levie