Postgres 19 is late because so many more bugs are being found now

	Postgres 19 is late because so many more bugs are being found now(rhaas.blogspot.com)
	595 points by soheilpro 2 days ago \| 363 comments
	tl;dr: PostgreSQL 19 is being delayed from its usual September/October release to November, with several major features (SQL/PGQ, UPDATE/DELETE FOR PORTION OF, partition MERGE/SPLIT) reverted late in beta. Back-patched commits jumped 60-74% year-over-year, driven by a flood of LLM-discovered bugs in the 30+ year-old codebase that committers must manually review and fix. The author suggests this foreshadows a long-term challenge: decades of latent bugs are now discoverable by AI, forcing Postgres developers (and software developers generally) to raise their standards.
	HN Discussion: ↑Finding and fixing latent bugs is a net positive worth the release delay ↑Firsthand confirmation that AI reveals an overwhelming number of real Postgres bugs ~Verification and maintainer review, not discovery, is the real bottleneck going forward ~Projects should triage bugs or rethink release cadence to handle the influx ↓The article oversimplifies; the reverts have more nuanced causes