Recently there was a call by thousands of Artificial Intelligence researchers to pause training of models more advanced than GPT4. We are getting to the point that what we are seeing is close to General Intelligence, and that there are very real risks.
Critics of the open letter have come in a few varieties. Some have claimed that the letter is scare mongering, and that we really have no evidence of the risk, while it may undermine progress and manifest benefits to humanity.
Others agree with the assessment of the risk itself, but don't believe the open letter will bring about a real moratorium. Even if there were a moratorium would it be long enough or effective in the long term? Or would be need a more severe ban on AI?
What is certain is that there is no certainty or agreement, even among experts in the field.
The video dialog is entirely generated by GPT4
Top comments (2)
Even if we were approaching general, or strong AI, why would a bunch of people saying, "let's hold off" have any effect on the people doing the research, who are all part of big businesses which aren't interested in ethical considerations beyond what happens to fall in line with making a profit?
My thesis was that it wouldn't. AI isn't a product that will be packaged and sold, but rather will be used by businesses to replace human labour. The race by the AI companies to be THE company is an existential one for them. There is no prize for second place. There is no reason to believe this won't follow a similar trajectory to other technology, and that those concerned with AI will be ignored. Further, as a strategy the open letter fails out of the gate.