I recently spent quite a bit of time with people outside the “tech bubble”.
Some are software engineers who work in slower-moving banking environments. Others are completely outside the tech industry.
The reactions to AI ranged widely. Some hate it, some are afraid of it, some don’t care, and others are excited about it.
Unsurprisingly, no one I spoke to is a power user. It’s hard when you’re on X, always testing the latest and greatest tools. However, everyone called out that AI makes mistakes, hallucinates, and is unreliable in different ways.
I believe the path into AI will keep getting more reliable. But that reliability will not come from the model alone. It will come from the foundation model, the harness around it, the data it can access, and the other layers people use to check its work.
Even then, I don’t know whether any individual’s definition of “perfect” will ever be achieved.
To explain this, I thought of an analogy on the spot, and it resonated:
Artificial Intelligence and Organic Intelligence have their own pros and cons. Neither is nor will be Perfect Intelligence.
The analogy did not make AI more reliable. It gave people a better way to think about its failure modes. That was enough to ground their expectations.
The conversation moved somewhere more useful after that.