Computer Science > Artificial Intelligence
[Submitted on 4 Sep 2025 (v1), last revised 15 Sep 2025 (this version, v3)]
Title:Murphys Laws of AI Alignment: Why the Gap Always Wins
View PDF HTML (experimental)Abstract:We study reinforcement learning from human feedback under misspecification. Sometimes human feedback is systematically wrong on certain types of inputs, like a broken compass that points the wrong way in specific regions. We prove that when feedback is biased on a fraction alpha of contexts with bias strength epsilon, any learning algorithm needs exponentially many samples exp(n*alpha*epsilon^2) to distinguish between two possible "true" reward functions that differ only on these problematic contexts. However, if you can identify where feedback is unreliable (a "calibration oracle"), you can focus your limited questions there and overcome the exponential barrier with just O(1/(alpha*epsilon^2)) queries. This quantifies why alignment is hard: rare edge cases with subtly biased feedback create an exponentially hard learning problem unless you know where to look.
The gap between what we optimize (proxy from human feedback) and what we want (true objective) is fundamentally limited by how common the problematic contexts are (alpha), how wrong the feedback is there (epsilon), and how much the true objectives disagree there (gamma). Murphy's Law for AI alignment: the gap always wins unless you actively route around misspecification.
Submission history
From: Madhava Gaikwad [view email][v1] Thu, 4 Sep 2025 23:03:25 UTC (375 KB)
[v2] Wed, 10 Sep 2025 14:35:38 UTC (370 KB)
[v3] Mon, 15 Sep 2025 06:39:32 UTC (21 KB)
References & Citations
export BibTeX citation
Loading...
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.