Meta scientist Yann LeCun says AI won't destroy jobs forever
This article from the BBC may be of interest. Here is a section:
"Will AI take over the world? No, this is a projection of human nature on machines" he said. It would be a huge mistake to keep AI research "under lock and key", he added.
What is AI and what risks does it pose?
Warning using AI for loans and mortgages is big riskPeople who worried that AI might pose a risk to humans did so because they couldn't imagine how it could be made safe, Prof LeCun argued.
"It's as if you asked in 1930 to someone how are you going to make a turbo-jet safe? Turbo-jets were not invented yet in 1930, same as human level AI has not been invented yet."
"Turbo jets were eventually made incredibly reliable and safe," and the same would happen with AI he said.
Meta Platforms made the decision several years ago to open source their AI data. In the aftermath of the Cambridge Analytica debacle the company probably reached the conclusion a hands off approach would defray risk. That decision helped to seed the AI industry of today.
It is an all too human attribute to think inanimate objects have human emotion or share our complicated ambitions. Investors do it all the time with stocks. Pet owners constantly refer to their dog or cat babies and we are in danger of doing the same with AI. The basic question comes down to intention. Large language models take the data they have been trained on and make a best guess at an answer. That does not mean it is correct but it often produces a solid draft.
The leap to general AI, where a computer can make independent decisions is not an iterative process. It will require a massive gain of function innovation that does not yet exist. Kudzu has an incredible growth rate but that does not mean it is self-aware. That in no way detracts from the productivity enhancements ChatGPT provides. It is a reason to stop fearing the imminent destruction of humanity and the performance of the respective stocks reflects that.
The most important thing to remember is acceleration is a trend ending. There is a rush to accumulate a position in their new sector and buyers are insensitive to price. Accelerations can go farther and faster than anyone believes possible but they do eventually soak up all available demand. At that point, all that are left are potential sellers. The first clear downward dynamics are likely to mark peaks of medium-term significance for Nvidia and Meta Platforms. Microsoft is less overextended but is not exactly cheap.