John McChesney-Young finished reading AI Snake Oil by Arvind Narayanan
From two of TIME's 100 Most Influential People in AI, what you need to know …
Fascinating non-technical study of AI, concentrating on three types: predictive, generative, and content moderation, going into details on current strengths (if any: predictive is very bad at it) and future prospects for them; there's also a section on Artificial General Intelligence. My only two criticisms are that although it addresses exploitation of workers in the Third World for the process of adding training materials, I think it underplays current inequalities caused by AI; and it does not mention at all the excessive use of water and power for the data centers AI uses. To be fair, the book may have been completed before the drastic resource needs were known, but it's unfortunate that the matter is absent, although perhaps a revised paperback version might include it. An errata note: figure 1.2 (p.29) lacks am application referred to on p.33, and the identification key to the images in figure 5.1 are …
Fascinating non-technical study of AI, concentrating on three types: predictive, generative, and content moderation, going into details on current strengths (if any: predictive is very bad at it) and future prospects for them; there's also a section on Artificial General Intelligence. My only two criticisms are that although it addresses exploitation of workers in the Third World for the process of adding training materials, I think it underplays current inequalities caused by AI; and it does not mention at all the excessive use of water and power for the data centers AI uses. To be fair, the book may have been completed before the drastic resource needs were known, but it's unfortunate that the matter is absent, although perhaps a revised paperback version might include it. An errata note: figure 1.2 (p.29) lacks am application referred to on p.33, and the identification key to the images in figure 5.1 are shuffled.
