Attendee Let's try to look at this from the point of view of data. I think history has a lot of examples of capital investment to improve productivity. For example, we do have much more efficient farms and manufacturing than we used to have. In some sense, this is an attempt to improve program development using automation; and just like it happened with the improvement of manufacturing capability, the change will draw natural attention to it.
Are critics suggesting there's too much too fast? Some of the data does suggest that. And a a result, we see examples of organizations downgrading their previous aggressive strategies (like the example in the presentation from Meta).
In situations where it's being judiciously applied, there is a bigger issue which I think is legitimate. It goes back to that very first statement that I quoted from Mark Zuckerberg, when he said AI could replace middle managers.
There is evidence that the value AI provides is very different for different levels of programmers. For expert programmers, it really does improve their productivity. They can use it to do auto-completion of code, and they can tell whether the generated code is correct because they have extensive experience working with good code. Whereas the junior guys don't have that increase in productivity.
That results in two effects. The first is a reluctance to hire the junior guys, because they just aren't as effective. But just as importantly, the AI is automating tasks that were previously used as mentoring, growth, and training tasks for the new guys. So there is concern about where the next generation of senior programmers will come from. If we're not training junior developers, or even hiring them, we may create a short term gain but a longer term problem.
Software Excellence Alliance