tuan
Joined 14 years ago
Joined 14 years ago
Every PR now has lots of unit tests, but they test the implementation details, not the spec. So now every change that breaks their implementation details causes false positive test failures. This creates a self enforcing negative loop. Every PR now comes with tons of unit test fixes.
People start responding to PR comments with something along the line of: I ask AI but it was not able to solve the problem, if you have a solution, LMK. Or another variant I see often is: I think this is wrong, but AI says this is fine, so I'll leave it as is.
I see craft lovers or product people using AI effectively. I use AI daily too. But the above camp is making my day to day job sometimes unbearable.
That maybe correct for some lessons. Many lessons you have to learn the hard way to really absorb them.
This reminds me of the day of Dreamweaver and the like. Everybody loved how quickly they could drag and drop UI components on a canvas, and the tool generated HTML code for them. It was great at the beginning, but when something didn't work correctly, you spent hours looking at spaghetti HTML code generated by the tool.
At least, back then, Dreamweaver used deterministic logic to generate the code. Now, you have AI with the capability to hallucinate...
Is it really new? We've been replacing real human connections with online connections/friendships for quite a while now. Social media companies have been giving us a world full of simulated relationships and making profits off of it. As quoted in the post, the average American adult has 3 friends. Look how many friends they have on FB.