- Joined
- Dec 1, 2018
- Messages
- 3,556
This is a long watch at around two and a half hours, but it is one of the more interesting and thought-provoking discussions I have seen about the current AI boom.
Ed Zitron challenges many of the claims being made by the AI industry, including the idea that generative AI will replace most jobs, deliver enormous economic growth and justify the trillions being spent on GPUs, data centres and infrastructure.
He also digs into the financial side of companies such as OpenAI and Anthropic, questioning whether their current business models can ever become sustainable without a major reduction in the cost of running these systems.
A major part of his argument is that the AI services people are using are heavily subsidised. He says the subscription price often does not cover the real cost of the computing being used, with the AI companies and their investors effectively covering the difference. His view is that current demand might look very different if users had to pay the full cost.
He believes the industry is being driven by hype, enormous amounts of subsidised investment and promises that the technology may not be capable of delivering. He predicts that
Ed Zitron challenges many of the claims being made by the AI industry, including the idea that generative AI will replace most jobs, deliver enormous economic growth and justify the trillions being spent on GPUs, data centres and infrastructure.
He also digs into the financial side of companies such as OpenAI and Anthropic, questioning whether their current business models can ever become sustainable without a major reduction in the cost of running these systems.
A major part of his argument is that the AI services people are using are heavily subsidised. He says the subscription price often does not cover the real cost of the computing being used, with the AI companies and their investors effectively covering the difference. His view is that current demand might look very different if users had to pay the full cost.
He believes the industry is being driven by hype, enormous amounts of subsidised investment and promises that the technology may not be capable of delivering. He predicts that
