Can AI run the world? Hell no! Let me count the reasons

 

1. AI doesn’t actually exist. At best, it’s a highly misleading misnomer: yes, it is artificial; no, it is not the least bit intelligent. A misnomer this misleading borders on fraud. AI is a pattern recognition/completion system. It cannot be intelligent because it lacks what phenomenologists call embodiment – the sum total of experience accumulated from spending a lifetime inhabiting a human body and being endowed with a sufficiently powerful brain.
2. AI grows more stupid over time. To start with, it lacks critical judgment and cannot distinguish human creations from the output of AI systems. Thus, the more it is used, the more error-prone it becomes. In the end, it becomes a noise amplification system.
3. AI systems cannot sense the limits of their, knowledge base. When asked an unfamiliar question, they spew random garbage.
AI systems may be successfully applied in certain limited domains. They are best used for functions normally performed by idiot savants.
Thus, the question reduces to this: Can idiots run the world? No, but they can run certain countries: the US, for instance, or France, or Germany, or Israel. Their leaders are idiots; not idiot savants but just plain garden variety ones. Look at Biden: bottom of his class at law school; never had a real job (US Congress and the White House don’t count). Let’s stop there because the rest of them are even worse.
So, don’t look for AI to run anything. It will find its niches in embedded systems where it will play purely advisory role. AI being trusted to make important decisions? Forget about it! Only fools would go for that.
Pete: No link because it’s behind a paywall. I will leave a clue though, it’s from a Russian blogger who lives in St Petersburg and who used to live in the US on a boat
Facebooktwitterredditpinterestlinkedinmail