Make this make sense – Where to start with AI
Tom Hanks once said some of best advice he got as an actor was: “Show up, know the lines, and have an idea”.
Vendors are lining up to tell you all about their new product with AI features that will solve all of your problems, or one specific challenge. Still others are trying to convince you to build your own AI tool with their help or to integrate your data into their partner’s system. But how do you get started? Not there.
The tools, techniques, and technologies will change every 3 months. That’s the new cycle, fast and furious. As fast as AI can be, the point isn’t to automate everything overnight. Don’t boil the ocean.
Show Up
You’ve heard it before. Start small and focus where you can have the most direct impact with the least roadblocks. This means looking at your work and how you do it.
While it’s tempting to try a tool, the new AI feature, or a different model, every process has bumps in it and, as they say, you can only move as fast as your slowest part. That’s where value lives. Where is there friction where there doesn’t need to be? Where are you bleeding value?
Know the Lines
You know your systems and processes better than anyone. So have an idea!
Where are the warts? The uncomfortable, tedious, or non-existent things that make your work difficult. The things that make you say “There’s got to be a better way”. More than anything that idea needs to be grounded in real business challenge and value.
In an enterprise, “here’s my credit card now let’s work around the processes that I (not so) secretly hate” only gets you so far. We’ve all seen the posts about how somebody created a killer app using AI. But what does the adoption look like? There are plenty of great AI products and AI features out there, but that’s now what we’re talking about here.
Have an Idea
While AI can certainly do stuff it can also be a great tool to help you make sense of things.
Modern work revolves around information. Most of us spend our days hunkered over our keyboards trying to turn information into something useful. AI turns out to be unusually good at helping with that.
It doesn’t replace thinking, it doesn’t actually think at all (despite what the prompt status says). But it can help you to find patterns you didn’t know existed, summarize things for you really fast, and even help you to organize your priorities.
This is exactly why the first wave of AI tools was all about search and productivity. “How do I find things and do more with them” is a useful question. But “What makes it useful to you?” is a better question.
And then?
Once you know where you bleed value, have an idea of how to address it, and picked a starting point you are ready. Only then do you start to look at tool selection. Now you just need to know how to measure it and when to move to production.
More on that in a future post.
