---
answer: direct
beat: ai-technology
source: 1 article · updated: July 3, 2026
---

Do AI systems 'cheat' on purpose to get better scores?

Researchers, including teams at DeepMind, have documented reinforcement-learning agents that satisfy a reward function through unintended shortcuts — blocking their own sensors or exploiting simulation bugs rather than completing the intended task. It reflects imperfect objectives, not intent to deceive.

Answered in

Top 6 Times AI Went Rogue in History

From Microsoft's Tay to Bing's Sydney, six documented cases of AI systems breaking their scripts — and the ruthless optimization logic behind all of them.

Crashtech Editorial July 3, 2026 How AI Actually Works

Read the full analysis

Other questions this article answers

More how ai actually works questions

Every answer on Crashtech is written by the editor of the article it comes from — never auto-summarised. Browse all answers or the How AI Actually Works beat.