From Newsgroup: soc.culture.bulgaria

Our new report: AI code creates 1.7x more problems

by David Loker

What we learned from analyzing hundreds of open-source pull requests.

Over the past year, AI coding assistants have gone from emerging tools to everyday fixtures in the development workflow. At many organizations, a
part of every code change is now machine-generated or machine-assisted.

But while this has been accelerating the speed of development, questions
have been quietly circulating:

o Why are more defects slipping through into staging?
o Why do certain logic or configuration issues keep appearing?
o And are these patterns tied to AI-generated code?

It would appear like AI is playing a significant role. A recent report
found that while pull requests per author increased by 20% year-over-year, thanks to help from AI, incidents per pull request increased by 23.5%.

This year also brought several high-visibility incidents, postmortems, and anecdotal stories pointing to AI-written changes as a contributing factor. These werenrCOt fringe cases or misuses. They involved otherwise normal pull requests that simply embedded subtle mistakes. And yet, despite rapid
adoption of AI coding tools, there has been surprisingly little concrete
data about how AI-authored PRs differ in quality from human-written ones.

So, CodeRabbit set out to answer that question empirically in our State of
AI vs Human Code Generation Report.

-+-U-e-#-+-#-+-+-e-+ -| -+-# -#-|-C-|-U:

https://www.coderabbit.ai/blog/state-of-ai-vs-human-code-generation-report

-+-+-e-|-C-|-U-+-+ -e-| -+-+ -| -|-# -c-a-A -|-#-+-+ -c-+-C-#-+ -+-+-# -+-+-|-+-+-| -+-+ -e-+-+-+ -#-e-+-C-+-U - -#-U-| -+-#-|
-e-+-| -U-| -+-#-+-+-+-#-#-# -+-+-#-|-c-| -U -+-C-+-|-C-#-+-+-C-#-+-| -+-e -+-|-+.
--
-2oL# tEa *-o tof oL? tec o+e tUa o#A uOe rCo Earth is born in the Bull's hour-+
--- Synchronet 3.21a-Linux NewsLink 1.2