4 ms·
First, often the premise of these pontifications conflate AI’s with automation. Automation will take many jobs, AI won’t. Second, the problem with these hypoth
by John23832 4y ago
First, often the premise of these pontifications conflate AI’s with automation. Automation will take many jobs,
AI won’t.
Second, the problem with these hypotheses is that AI’s don’t really work better than humans. They just are cheaper and corporations are in the position that they can force customers to use them instead of humans.
Nobody wants to call an AI model . They would rather call a human with the ability to actually solve their problems. When you call an AI, you don’t have something that can solve your problem, you’re presented with a stone wall of options (most of which are suboptimal). The same goes for most direct human to AI interactions.
AIs (short of a true general AI) work best in augmenting human ability. Helping doctors/radiologists read X-rays and mris better. Helping farmers predict weather patterns for farming. Folding proteins. These are the best utilizations of AI. Not wholesale replacements or humans.
- ChuckNorris89 4y ago>They would rather call a human with the ability to actually solve their problems. Unfortunately a lot of humans in customer support are also failing to solve my problem since they have little say in the system, and are mostly meat based automatons following a script or logic flow chart they've been given. If your problem doesn't fall into what they've been taught to do, well, then tough shit for you, you might as well have been talking to an AI bot instead of the metatag getting paid minimum wage to take shit from customers angry at the corporation for failing to provide the service they took your money for.
- John23832 4y agoTrue, but I think it’s easier to solve the meat problem than the silicon one.
- JohnFen 4y ago> Automation will take many jobs, AI won’t. AI _is_ automation.