AI isn't a feature of your product

your product is a feature of AI

(The idea for this comes from a conversation with Manik Gupta this week, who said I could steal it!)

As folks wake up to the potential of AI across the industry, we are seeing a common pattern: people “adding some AI” to their product or process. But, as the title of this post says, that might not necessarily be the right way to think about it.

In general, as new paradigms emerge, we think in terms of familiar patterns. In the early stages of the internet, companies that had been building desktop software took desktop-centric approaches to the internet and built things like web page editors, browser plugins, etc. But the real energy and value came from treating the internet as a platform. Amazon, eBay, Google, Facebook don’t have “some internet as a feature of their product”. Their products are fundamentally built for the internet. They don’t have the internet “mixed in” - for each of those products, there is no product without it!

I believe that AI is, or will become, another platform at the scale of the internet. It’s early stages now, and there are plenty of challenges and technical issues to be dealt with, but setting that aside for a moment, the same pattern is very likely to hold here. It will be possible to add some value by “building AI into your product” but really transformative, massive value will come from building apps and solutions that won’t work at all without it - that treat it as a true platform.

We don’t know the shape of those apps yet, and we are barely at the point where we understand what the platform is, how to use it, and what best coding practices even are. All of this was true at the start of the PC, Internet, and Smartphone platforms too.

As always, this is a good time to ask the “what if” questions, not the “why not” ones. What if this really is a new platform? Assume it all works well - what would you build?