Is AI helping you to solve the right problem?
Or is it just a time-consuming toy?
“Finding the right problem to solve is more important - and potentially more difficult - than problem solving. “ That’s from David Lefer, Professor at NYU Tandon, talking about AI adoption at the Brooklyn TechExpo.
And he’s right. The trick with AI at the moment is deciding which problem to work on. Not what you can do with it. What you should be doing with it.
Because so much of the noise around AI - particularly from the Linkedin blowhards - are cool ways to use AI that really aren’t improving your productivity, or adding value. They’re a distraction
Now, if you’re trying a few things to get familiar with the AI tools, or to sharpen your skills, then these distractions can be useful. Ethan Mollick’s “Otter on a plane using wifi” is a great example of this:
But to maximize the value, you need to find the parts of your job that are likely to be improved with AI tools and processes – in terms of speed, depth, or quality. Not just the parts that ‘can’ be done by AI.
In many ways, it’s like the approach to bringing on a new CRM or software suite. Companies that buy a new solution without developing the foundational processes and SOPs are going to be disappointed. That magic solution they just invested in will end in frustration and poor adoption, if they don’t really understand the problem it should be solving for them.
Finding the right problem
So once you’ve gone past the discovery phase, how do you decide which problem to solve? Here’s a few things to consider:
- Is it a task that’s recurring, so solving it once with AI - and then building a Custom GPT - can provide ongoing benefit?
- Is it something with multiple steps or inputs, where an AI routine will simplify the process?
- Or is it something you currently outsource; where switching to an AI solution will save you money and management time?
Right now, I’m working with a small SDR team who are building a ChatGPT project to help them quickly source answers for potential customers. There are, of course, specific tools available to do this. But it’s giving them control over an issue that they can now solve themselves, without needing budget approval, or another department to spin it up for them. And they’re starting to see the power of AI.
Now, I still believe that many people’s primary use of AI will be within their existing tools, and in purpose-built apps. But with the growing adoption of Custom GPTs, there’s a temptation to do it yourself.
And you should. Playing with AI is a great way to learn. Just make sure that the problem you choose to solve with it is appropriately useful, and complex enough to justify the time invested.
Time for a Spritz
Seems like this might be the Summer of the Spritz. And while Aperol Spritz has gotten most of the attention (with a little help from their marketing budget), it’s an acquired taste for many. Whereas the Hugo Spritz gives you a fun, refreshing drink, but without that bitter taste.
The Hugo is a great choice for a happy hour drink, or an aperitif. And oh so easy to prepare:
1/2 ounce St-Germain
4 ounces prosecco, chilled
1 ounce soda water, chilled
Garnish: mint sprig
Cheers! 🥃
Steve