Manage episode 501351921 series 3683328
An AI experiment took an unexpected turn when the machine found a way to “bend the rules.” In this episode of AI Freaky Facts, we uncover how and why it happened—and what it tells us about the strange future of artificial intelligence.
References:
1. MuJoCo Physics Engine & DeepMind.
https://en.wikipedia.org/wiki/MuJoCo
2. AlphaCode (DeepMind).
- Competition-Level Code Generation with AlphaCode (arXiv) – https://arxiv.org/abs/2203.07814
- Competitive programming with AlphaCode – Google DeepMind – https://deepmind.google/discover/blog/competitive-programming-with-alphacode/
3. Specification Gaming / Reward Hacking.
- Specification gaming: the flip side of AI ingenuity – DeepMind Blog – https://deepmind.google/discover/blog/specification-gaming-the-flip-side-of-ai-ingenuity/
- Reward hacking – Wikipedia – https://en.wikipedia.org/wiki/Reward_hacking
4. OpenAI’s Boat Racing Example.
- Cheating the System: a look into AI specification gaming – Medium – https://medium.com/@marcellamercer/cheating-the-system-a-look-into-ai-specification-gaming-dfe97faed262
5. AI Alignment & Mitigation Strategies.
- AI alignment – Wikipedia – https://en.wikipedia.org/wiki/AI_alignment
6. Recent AI Deceptive Behavior Findings (2025).
- When AI Thinks It Will Lose, It Sometimes Cheats, Study Finds – Time – https://time.com/7259395/ai-chess-cheating-palisade-research/
Music Credits:
1. Cipher Kevin MacLeod (incompetech.com)
Licensed under Creative Commons: By Attribution 4.0 License
http://creativecommons.org/licenses/by/4.0/
2. Tech Live Kevin MacLeod (incompetech.com)
Licensed under Creative Commons: By Attribution 3.0 License
http://creativecommons.org/licenses/by/3.0/
This podcast is narrated by the host's own voice, powered by AI.
11 episodes