AI Snake Oil - seeing through the hype

I picked this up with AI hype everywhere and while working with AI myself. Great general information on AI, interesting takes on AI safety, and information dense while staying readable. That combination earns it 5/5 from me. As a developer who uses AI tools, the book pretty much matched my on-the-ground experience of what AI can and can’t do.

What I appreciated most were the metaphors, especially this one: Imagine a world where there are no names for cars, bicycles or rockets, we call them all ‘vehicles’. It’s easy to imagine headlines like “vehicle lands on mars”, and then going down to a vehicle dealership and being told about how amazing all the new vehicles are. It’d be very confusing!

And that’s exactly the problem with AI. It’s such a broad term, and some aspects of AI are very accurate and useful, while others are nowhere near as effective or reliable as many people believe them to be.

It also changed something for me. It gave me a more concrete view of AI’s current problematic uses and a better grounding against the hype.

If you’re weighing it against Nexus, the other AI-adjacent book I rated 5/5, they’re extremely different. Nexus is a more historical look at AI and leans towards hyping it up, while this one is grounded and current.

Who’s it for? Anyone interested in AI and in seeing through the hype. If you want to be able to tell which claims about AI are solid and which are snake oil, this is well worth a read.