Public Backlash Against AI Is Real: Violence, Polling Data, and Diminishing Returns
On April 10, a 20-year-old threw a Molotov cocktail into Sam Altman's home. Days earlier, someone shot 13 rounds into a councilman's home in Indianapolis over a data center project. These are the extreme edges of a growing public backlash against AI, but the subtler data is just as stark. Stanford's 2026 AI Index reports that only 23% of the public think AI will improve jobs long-term (vs 73% of experts), and only 21% think it will help the economy. A Gallup poll from March 2026 shows Gen Z's excitement about AI dropped from 36% to 22%, while anger rose from 22% to 31%.
Why the backlash is accelerating
Tech journalist Jasmine Sun describes the mood as a worldview where AI is seen as an elite political project pushed by out-of-touch billionaires. Meanwhile, AI CEOs oscillate between apocalyptic warnings and promises of mass job displacement, while the public faces rising costs and stagnant wages. In Virginia, data center buildout could raise residential electrical rates by 25% by 2030.
The productivity numbers don't help
A February 2026 NBER paper found 80% of companies using AI reported no impact on productivity. A widely cited 2025 MIT study showed 95% of corporate AI pilot programs got zero return. Even in coding, ML engineer Han-Chung Lee argued on GitHub that internal productivity metrics are often skewed to hit adoption targets nobody can audit.
📖 Read the full source: HN AI Agents
👀 See Also

Constraint Decay: Why LLM Agents Fail at Structured Backend Code
New research introduces 'constraint decay': as structural requirements accumulate, LLM agent performance drops drastically — capable agents lose 30 points in assertion pass rates, weaker ones approach zero. Actionable insights for anyone using AI coding agents.

AI Tools May Lead to Homogenized Output in Creative and Development Work
A Reddit user reports that multiple teams using AI tools like ChatGPT, Co-Pilot, and Claude for strategy roadmaps and software development are producing similar outputs with identical buzzword patterns and design structures.

Grammar-Based Method Matches or Outperforms AI in Authorship Analysis
A University of Manchester study found that LambdaG, a grammar-based authorship analysis method, matched or exceeded leading AI systems across most test datasets while offering greater transparency and lower computational cost.

Claude Code v2.1.132: SIGINT Graceful Shutdown, MCP Fixes, and Terminal Handling Overhaul
Claude Code v2.1.132 fixes graceful shutdown on external SIGINT, adds CLAUDE_CODE_SESSION_ID and CLAUDE_CODE_DISABLE_ALTERNATE_SCREEN env vars, patches MCP memory leaks and tool listing retries, and resolves dozens of terminal edge cases across IDE terminals.