‘सामान्य’ ए.आई. भविष्य के पक्ष में तर्क | द एज़रा क्लेन शो

1위@usmcnomad5 दिन पहले

Member of the cybersecurity community here: we are not saying these are solved problems, we are actively and vocally telling everyone who will listen that these are not solved problems and are aggressively pushing for new standards, frameworks, protocols, tactics, and strategies explicitly because that innovation is necessary. Where we are attempting to apply existing approaches to this problem it’s because we are using what is available to mitigate risks to the extent that we can since our organizations are not slowing down to wait for the innovation.번역 보기

3위@JohnnyBravo682606 दिन पहले

Wow, cool, my unaligned AGI is checking on my other unaligned AGI. How could this possibly go wrong? Alignment is an unsolved problem no matter what this guy says. It's not an engineering problem; it's a science problem, and we don't even know if it's possible in principle, let alone in practice. AI checking AI is just hand-waving it away.번역 보기