Open-World Evaluations for Measuring Frontier AI Capabilities
Benchmark-based evaluation remains important for tracking frontier AI progress, but it can both overstate and understate deployed capability because it privileges tasks that can be precisely specified, automatically graded, easy to optimize for, and run with low budgets and short time horizons. This paper advocates for a complementary class of evaluations, termed open-world evaluations: long-horizon, messy, real-world tasks assessed through small-sample qualitative analysis. The authors introduce CRUX, a project for conducting such evaluations regularly, and demonstrate an AI agent successfully developing and publishing an iOS app with only one manual intervention.
[2605.20520] Open-World Evaluations for Measuring Frontier AI Capabilities
[Submitted on 19 May 2026]
Title:Open-World Evaluations for Measuring Frontier AI Capabilities
View a PDF of the paper titled Open-World Evaluations for Measuring Frontier AI Capabilities, by Sayash Kapoor and 17 other authors
View PDF HTML (experimental)
Abstract:Benchmark-based evaluation remains important for tracking frontier AI progress. But it can both overstate and understate deployed capability because it privileges tasks that can be precisely specified, automatically graded, easy to optimize for, and run with low budgets and short time horizons. We advocate for a complementary class of evaluations, which we term open-world evaluations: long-horizon, messy, real-world tasks assessed through small-sample qualitative analysis rather than benchmark-scale automation. In this paper we survey recent open-world evaluations, identify their strengths and limitations, and introduce CRUX (Collaborative Research for Updating AI eXpectations), a project for conducting such evaluations regularly. As a first instance, we task an AI agent with developing and publishing a simple iOS application to the Apple App Store. The agent completed the task with only a single avoidable manual intervention, suggesting that open-world evaluations can provide early warning of capabilities that may soon become widespread. We conclude with recommendations for designing and reporting open-world evals.
Subjects:
Artificial Intelligence (cs.AI)
Cite as: arXiv:2605.20520 [cs.AI]
(or arXiv:2605.20520v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2605.20520
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Stephan Rabanser [view email] [v1] Tue, 19 May 2026 21:42:32 UTC (1,131 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Open-World Evaluations for Measuring Frontier AI Capabilities, by Sayash Kapoor and 17 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.AI
new | recent | 2026-05
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Loading...
Data provided by:
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
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.
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)