"The problem is that code produced by an AI agent looks reasonable, but can contain ‘hard-to-spot bugs’ that end up causing major problems. As a result, you should carefully review your agent’s output, but this is difficult. It’s “famously hard” to understand code you didn’t write yourself, so this extra step becomes “easy to just blow it off”. Soon, systems start to break." -- Cal NewportFrom "On AI Coding and Its Discontents":https://calnewport.com/on-ai-coding-and-its-discontents/#llm #softwarequality #cognitivedebt
Related
@mariyadelano > I’ve been testing a theory: many people who are high on #AI and #LLMs are just new to automation and don...
@mariyadelano > I’ve been testing a theory: many people who are high on #AI and #LLMs are just new to automation and don’t realize you can automate processes with simple programmin...
One Pepper, Every Door I built Pepper’s voice features with one customer in mind: my dad. Reading a screen isn’t always ...
One Pepper, Every Door I built Pepper’s voice features with one customer in mind: my dad. Reading a screen isn’t always easy for him, and I wanted him to be able to just talk to Pe...
The Poincare conjecture in HOL Light, formalized by Claude Opus 5.5 following Perelman. ~ John Harrison. https://github....
The Poincare conjecture in HOL Light, formalized by Claude Opus 5.5 following Perelman. ~ John Harrison. https://github.com/hol-light-family/Poincare #HOL_Light #AI4Math #AI #Math