Why LLM writing is tiring: It mimics software engineering rigor
ipeirotis · x · 2026-08-27
Panos Ipeirotis theorizes that the default LLM writing style is exhausting because models are heavily optimized for software engineering, which leaks into their text generation. Every sentence tends to be correct, hedged, and defensible, much like production code. This results in prose that is overly precise, robust, and filled with unnecessary details.
Related event: Why LLM Writing Feels Exhausting: Software Engineering Bleeds In(2 posts)→
More from Research
- TheoPhysEval paper: GPT-6 Astra solves research-level physics problems at 98% accuracy — alex_verem · 2026-10-16
- GPT-6 Astra scores 98% on expert-crafted theoretical physics benchmark, solving day-long problems for under $3 — alex_verem · 2026-10-16
- Researchers propose public desk-rejection records and submission bans on OpenReview — tallinzen · 2026-10-16
- WEDGE uses navigation agents' own failures to generate targeted training environments — EliasEskin · 2026-10-16
- HYSET scores tool sets instead of single tools, lifting agent task success to 69.9% — DeepLearningAI · 2026-10-16
- Microsoft's Magnetar zkVM proves programs 10-39x faster using only standard hash functions — Microsoft Research · 2026-10-16