It’s better to be in AI now than it was a year ago (if you care about building great products)
Image: “PD001: Figure 0.1” by Rosenfeld Media is licensed under CC BY 2.0
The bloom is off the rose for AI these days:
📰 The tech news is filled with questions about financial viability and consumer disillusionment.
😩 AI conversations with colleagues start with sighs and throwing up of hands.
🚫 The once-miraculous, near-human responses of ChatGPT have become mundane, recognizable–even grating.
Some analysts argue we’re entering the “Trough of Disillusionment,” which sounds bad (and probably is, if you’re an investor).
But if you build products and services, it’s more like the dust is finally settling.
What’s valuable about where we are now is that this is where signal improves, and it becomes vastly easier to recognize where there are real human needs that can be addressed with AI.
One of the biggest obstacles to validating emerging technology applications is the novelty effect. When users are dazzled that a computer has just talked to them, it’s impossible* to get meaningful data about whether this could improve their daily lives (*at least, from them, directly).
But when the average consumer has enough experience with LLMs to see the problems with Google’s “Dear Sydney” ad, that says something important. It means that our users are able to provide more discerning feedback on how LLMs should show up in their own lives.
In our AI fatigue, it’s also easy to overlook that “AI” is too big to sit squarely at any one stage of the hype cycle. LLMs and image generation may be entering the disillusionment stage, but other types of AI–such as audio/video generation, physical AI, possibly new smart glasses–have yet to reach their peak. Meanwhile, other types of machine learning have been in the “Plateau of Productivity” for years.
So while we all have good reason for feeling burned out, the reality is that we are in a time of opportunity. If your goal is to design great products and services that solve real problems, it’s becoming easier to do that with AI each day.
For my part, I’ve been through this many times before, both in house, and now, in my consulting work.
My work with teams is all about cutting through the hype of new technologies to find real human problems worth solving. I’ve guided teams through the process of building evidence-based business models for many emerging technologies, before, during, and after the peak of the hype cycle. (If you’re looking for help figuring out your users for an innovative product or service, we should talk. 👋 )
From this perspective, I can attest that the best cure for technology fatigue is to solve meaningful problems. And with AI, and we are in a better position to do that now than we have been since ChatGPT was introduced.