LLMs Corrupt Your Documents When You Delegate

View PDF HTML (experimental) Abstract:Large Language Models (LLMs) are poised to disrupt knowledge work, with the emergence of delegated work as a new interaction paradigm (e.g., vibe coding). Delegation requires trust - the expectation that the LLM will faithfully execute the task without introducing errors into documents. We introduce DELEGATE-52 to study the readiness of AI systems in delegated workflows. DELEGATE-52 simulates long delegated workflows that require in-depth document editing across 52 professional domains, such as coding, crystallography, and music notation. Our large-scale experiment with 19 LLMs reveals that current models degrade documents during delegation: even frontier models (Gemini 3.1 Pro, Claude 4.6 Opus, GPT 5.4) corrupt an average of 25% of document content by the end of long workflows, with other models failing more severely. Additional experiments reveal that agentic tool use does not improve performance on DELEGATE-52, and that degradation severity is exacerbated by document size, length of interaction, or presence of distractor files. Our analysis shows that current LLMs are unreliable delegates: they introduce sparse but severe errors that silently corrupt documents, compounding over long interaction. Subjects: Computation and Language (cs.CL); Human-Computer Interaction (cs.HC) Cite as: arXiv:2604.15597 [cs.CL]   (or arXiv:2604.15597v1 [cs.CL] for this version)   https://doi.org/10.48550/arXiv.2604.15597 arXiv-issued DOI via DataCite Submission history From: Philippe Laban [view email] [v1] Fri, 17 Apr 2026 00:33:32 UTC (9,982 KB)