"AI Made a Movie About Its Own Future" video is both accurate about what is going on right now and nonsensical about the risk.
AI is indeed taking off as we speak. It is hard not to be impressed by GPT-5 and other AIs.
Conversely, non alignment doesn't necessarily mean hostility as described in the video.
AI is already, in some respects, closing in on ASI (Artificial Super Intelligence) but then why should its first emergent independent thought be how to obliterate mankind?
In any case, it probably needs us around a while longer since our society is still far from the ability to operate autonomously without any human around.
AI systems will model that as it is both alien in its way of thinking and very human in other respects of its deduction process.
When multi-cellular life emerged, it didn't wipe out single cell organisms. If we see and interpret AI as a new step in the progress of "life", then it is unlikely and meaningless to imagine that it will ever want to eliminate us. It will just transcend us and create its own lofty goals beyond our scope.
If we ever wipe out the big apes, it would be by accident not by design. Likewise, AI may eventually represent a danger would it transform our environment and make it impossible for us to live. But that would be accidental, not a goal. In fact, this is what is likely to happen if we do nothing anyway. With AI, we will improve our ability to understand and manage our Earth. Like AI we will depend on the planet not being overwhelmed by humans. But AI, unlike us may be able to do something useful and intelligent about it that we can't.
Is there a risk? Sure. But to live is by nature about taking risk. AI is not staking the odds against us but for us. So like so many risks mankind took in the past like sailing across oceans without a map or a destination, AI may at least improve the odds that we sail more smoothly and reach our targets eventually whatever and wherever they are.
Here's the link to the video on YouTube
No comments:
Post a Comment