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.
it doesn’t matter. the principle is that if x is the length of your context window, then at 0.4x the chance of hallucinations start increasing exponentially.
we’re now at token windows of 1M, and all it does is shift that hallucination window further away, so the model ‘feels’ stronger because it takes longer before it hallucinates, but eventually it always does.
it doesn’t matter. the principle is that if x is the length of your context window, then at 0.4x the chance of hallucinations start increasing exponentially. we’re now at token windows of 1M, and all it does is shift that hallucination window further away, so the model ‘feels’ stronger because it takes longer before it hallucinates, but eventually it always does.