They see their competitors as Claude Code, Codex, and Cursor, not ghostty or something.
> writing code is now fast, it's getting cheaper, and quality is going up to the right
My hobby projects have 100x more tests than they used to, because LLMs are great at writing tests. And my subjective experience is that the net quality has increased as a result.
YMMV, but it’s certainly a common belief, and for me at least a lived experience.
I don't care what people believe, I care about what is. What is measurable. What is factual. I need evidence for a belief to be meaningful. I need strong evidence for a belief to be strong. Not just evidence in favor, but evidence that alternative explanations are unlikely.
Currently I see evidence that things are moving fast. But I am unable to distinguish if this is actually because of AI or because increased efforts and motivation. Most importantly, speed isn't the same thing as velocity.
What I do not see evidence for is increasing quality. In fact, I see strong evidence to the contrary. I see strong evidence that quality is declining even quicker than it was before. I'm not convinced AI is that cause of this, but there's more than adequate evidence for me to believe it is a (significant) catalyst.
Right now you could throw a stone in a random direction and there's a good chance that whatever it lands on will be decreasing in quality. It is even easy to gesture broadly at Microsoft, but they aren't the only big tech disappointing their users.
I hear a lot of claims. I don't see a lot of evidence.
For my part, I’m too busy building to spend time quantifying quality issues in my past decade of hobby projects so I can compare to my new work, which subjectively is far more capable and higher quality.
I welcome the experimentation, there will definitely be something new, but this ain't it. New primitives are needed, at a higher level of conceptualization, not merely a fancy new interface.