Ask HN: Is it just me, or is software buggier across the board?

Hacker News Discussion ↗

Many users report that everyday software feels noticeably buggier and less reliable, citing broken core features in major products from Microsoft, Google, Apple and others, as well as flaky web and mobile apps. Explanations range from growing system complexity, “ship fast” culture, diminished QA and perverse business incentives, to the rising use of AI-generated code and “minimum viable” releases. A minority counter that core infrastructure like Linux and foundational libraries are more stable than ever, suggesting the decline is concentrated in consumer-facing, corporate-controlled services.

Overall Perception of Bug Trends

  • Many participants feel software is noticeably buggier, especially in the last few years; some say they now expect critical bugs in everyday tasks.
  • Others argue software has always been buggy; what changed is type of bugs (from OS crashes to UI/UX glitches).
  • A minority report no noticeable decline, or even improvements (especially on some Linux desktops).
  • Several note it’s hard to separate real decline from confirmation bias and changing expectations.

Platform- and Product-Specific Observations

  • Frequent complaints about Windows 11 (Explorer, updates, drivers, BitLocker/boot issues) and Microsoft 365 web apps (PowerPoint, OneNote, Word in browser).
  • iOS and Android updates are perceived as more fragile; users report UI elements not appearing, calls not working until reboot, and “scared to update.”
  • Google products (Drive, Sheets, Meet, YouTube, Google Home) and big web apps (ticketing, events, Enterprise rental site, HSA portals, Visa, eBay, Slack, Discord, Target) cited as increasingly unreliable.
  • Some gaming examples: formerly stable games becoming buggy after recent patches; AAA titles and big studios criticized.
  • Contrast: several users report Linux desktop and core tools (ffmpeg, virtualization, some DEs) are more stable than ever.

Proposed Causes

  • Feature velocity, “release early/often,” MVP mindset, and cheap updates leading to 80%-done products.
  • Systemic underinvestment in QA; QA teams cut or offloaded to developers and customers.
  • Management incentives favor new features, redesigns, and “busyness” over polish and regression fixing.
  • Growing complexity: more platforms, devices, libraries, threads, and integrations make systems fragile.
  • “Enshittification” and late-capitalist incentives: profits prioritized over craftsmanship and backward compatibility.

Role of AI

  • Some strongly link increased bugginess to AI-assisted coding and “vibe coding,” claiming code is shipped with less review and weaker tests.
  • Others say AI is not uniquely to blame; poor leadership and inadequate test coverage would produce bad outcomes with or without AI.

Measurement and Unclear Points

  • Several note the discussion is anecdotal and call for objective indices (e.g., trends in errors, issues, complaints).
  • Overall magnitude and timing of any “collapse” remain unclear from the thread alone.