2026 tech layoffs reach 45,000 in March

Original Article ↗ Hacker News Discussion ↗

Meta, VR, and AI Strategy

  • Meta is rumored to be planning another large layoff (around 20%), with some expecting only data center and infrastructure roles to remain core.
  • Many see Meta’s post-Instagram bets (Metaverse/VR, AI chatbots, social spinoffs like Threads) as financially underwhelming or outright failures, even if technically ambitious.
  • There’s debate on whether Meta is innovative: some cite strong R&D, open-source work (e.g., compression, kernel I/O, hardware projects), and Meta Glasses as genuinely good; others argue the company mostly copies or acquires rather than invents.
  • Llama is viewed as technically impressive but strategically mishandled: it should be a top-tier model but is seen as less capable and/or less effectively productized than competitors.
  • Meta’s core ad/surveillance business remains very lucrative, but many criticize it as ethically dubious and addictive rather than socially beneficial.

AI: Cause of Layoffs vs Excuse

  • A major thread argues layoffs are primarily a cyclical correction after years of zero-interest-rate policy, COVID overhiring, and investor pressure, with “AI” used as a PR cover.
  • Others point out companies explicitly tying cuts to AI capex and “AI-assisted efficiency,” reallocating money from staff (OPEX) to data centers, chips, and power (CAPEX).
  • Some report real productivity gains from AI tools (2–3x output for a single frontend dev), making it harder to justify larger teams in the short term.
  • Several expect Jevons-like effects long-term (more software, more demand) but see current cuts as a knee-jerk, short-sighted response.

Bloat, Metrics, and Organizational Health

  • Many claim big tech could cut ~20% of staff with minimal immediate impact due to layers of “process/meeting people” and long-accumulated bloat.
  • Others warn that simple headcount cuts without fixing underlying processes can worsen operations, especially when “duct tape” roles are removed without fixing core systems.
  • Attempts to quantify engineer productivity via metrics (lines of code, tickets, etc.) are criticized as easily gamed and subject to Goodhart’s law; such systems may select for visibility and metric-gaming over real impact.

Worker Experiences and Market Conditions

  • Laid-off workers report ghosting after multiple interviews, difficulty finding even non-tech jobs, and suspicion that a recession may be starting.
  • Some recount saving their employers significant sums yet still being cut, reinforcing the belief that performance doesn’t strongly protect against layoffs.
  • Career advice from the thread emphasizes perceived value and visibility to management over actual output, and warns that job security is inherently fragile.
  • A few are pivoting to building their own products, betting that higher-quality hand-crafted software and non-SaaS models can compete with AI-assisted “cheap” output, though success is uncertain.

Broader Structural Views

  • Several see the “money tree” era as over: companies must now choose between expensive GPUs and humans, and often choose GPUs.
  • Some contend the “age of SaaS” and easy software money is waning, with many roles revealed as unnecessary in hindsight.
  • Others argue that, beyond hype, AI agents are not yet meaningfully replacing human jobs; instead, tighter money and macro conditions are driving cuts, with AI mostly reshaping where capital flows.