The trouble with technology, of any kind, but certainly here, isn't the technology itself, but humans trust in the technology. If doctors use ChatGPT as a sanit

The trouble with technology, of any kind, but certainly here, isn't the technology itself, but humans trust in the technology. If doctors use ChatGPT as a sanity check of "can you think of anything i didn't?" and then ignore it from then on, it would be a good tool. But pretty soon people tend to change their perspective and say, well, ChatGPT would know...., so,... i'll go with it. As a developer, i'm pretty interested in static and dynamic code analysis as a way to easily find bugs, and it does do this pretty well. If developers use it as a tool to use as a reason to walk through code and examine it yourself, it is really quite powerful. It seems invariably, however, that people start trusting what the analysis tool says, and don't question whether the recommendations are correct or worth it. It's a powerful cognitive effect that would be interesting to study, that probably happens with all kinds of tech. Some are more dangerous than others.