AI and Everything Else - Benedict Evans | Slush 2023

Exploring the reality of AI, beyond the hype and focusing on its practical applications, automation, optimization, and convenience, and the three stages of platform shifts.

Key takeaways
  • AI is not just about intelligence, it’s about how it’s used, and it’s often used for automation, optimization, and convenience.
  • The tech industry is obsessed with what will happen in 2030 and AI, but it’s not clear what that will look like.
  • Platform shifts tend to go through three stages: excitement, disillusionment, and then a gradual adoption.
  • The current hype around AI is similar to the hype around VR in the 1980s and 3D TV in the 2000s.
  • AI can be used to automate tasks, but it’s not always clear what the right answers are.
  • The idea of general intelligence (AGI) is still unclear and it’s not clear what it would look like or how it would be achieved.
  • Large language models (LLMs) are getting a lot of attention, but they are not necessarily a step towards AGI.
  • The challenge is to work out the right level of abstraction for AI, and it’s not clear what that will look like.
  • The tech industry is often focused on what’s new and exciting, but it’s not always clear what the practical applications will be.
  • AI can be used to solve specific problems, but it’s not always clear what the broader implications will be.
  • The current hype around AI is driven by the excitement of seeing what’s possible, but it’s not clear what the long-term consequences will be.