The paper studies how to curate the fixed context files that large language model agents load at the start of each session. It models this as a capacitated assortment problem, proving bounds on optimal file size and showing that simply appending all positive‑value instructions can be far suboptimal. The authors analyze censored feedback from past sessions, derive evidence‑collection thresholds for adding instructions, bound regret under limited human review, and empirically demonstrate that irrelevant rules in real context files hurt model compliance.
By Zexuan Liu, Yuning Yang, Tiancheng Zhao
The paper introduces the concept of "empty commitments," promises made by AI agents that cannot be fulfilled due to the absence of necessary tools or runtime capabilities. It defines three failure types, proposes a detection checker, and evaluates it on 400 human-labeled examples and a benchmark of 293 follow‑up requests. The study shows that many open‑weight models frequently issue such empty commitments, and that simple interventions—informing the model of its runtime, adding a capability card, or running a checker loop—can significantly reduce these failures.
By Jiaqi Tang, Bingyu Shen, Lan Wei, Qing Lu, Bethel Ololade, Danny Galvis, Miles Q. Li, Bin Hu, Boyang Li
The paper introduces Verifiable Agent Insurance (VAI), a framework that lets AI agent providers commit to policy compliance through smart contracts and escrow funds on a blockchain. When a user detects a policy violation, a verifier reviews execution traces logged by the agent harness; if the violation is confirmed, the escrow is paid to the user. Experiments on three agent benchmarks show VAI detects 42.2% of violations, rising to 83.0% when agents emit explicit structured records.
By Cheng Hua, Taro Tsuchiya, Nicolas Christin, Giulia Fanti
The paper introduces the concept of an irreversibility budget for managing risk in fleets of large language model agents. It proposes a cumulative accounting system that tracks residual value‑at‑risk across agents, workflows, and tenants, charging each effect its residual loss and denying any marginal effect that would exceed the budget. A controlled study shows that per‑effect gates can allow fleet‑level overdraws up to 48 times the tenant’s risk limit while the budget keeps all correctly charged runs within limits, highlighting the challenge of designing conservative, dependency‑aware pricing.
By Bardia Mohammadi, Laurent Bindschaedler
The paper introduces the Two-Ledger Model to address the mismatch between settlement‑grain billing data and transaction‑grain cost accounting in multi‑agent AI systems. It proposes a settlement ledger of FOCUS‑conformant billing data linked to an attribution ledger of invocation‑grain usage records built on OpenTelemetry GenAI telemetry, using a layered correlation spine to map each record to its business transaction. The model is validated by a simulation that demonstrates high cost coverage, accurate residual detection of unmetered spend, and significant cost reductions through mid‑flight gating.
By Vivek Acharya
arXiv:2606.00341v2 Announce Type: replace-cross
Abstract: As AI agents are increasingly deployed in real personal and corporate settings (email accounts, development workflows, company databases, etc...
By Jeremy Tien, Abishek Anand, Yu-Rou Tuan, Yuchen Shen, J. Zico Kolter, Aran Nayebi
arXiv:2608. 02670v1 Announce Type: cross Abstract: Coding agents increasingly run inside organizations whose security controls (scoped credentials, restricted egress, read-only filesystems, non-root execution) constrain them like any other software.
By Dotan Davidovich, Yair Amar, Hai Rozencwajg, Or Hiltch
The paper introduces a formal framework for agent harnesses that guarantees termination, prevents drift, and enforces spend limits through bounded loops, gates, and repair relations. It proves that these guarantees hold even with repair budgets and demonstrates the effectiveness of the system by identifying vacuous gates and achieving low false‑accept rates in a 69‑loop catalogue. The authors provide an instrumented implementation and a held‑out mutant corpus to validate gate correctness.
By Varun Pratap Bhardwaj, Garima Singh, Arun Pratap Bhardwaj
The paper discusses how enterprises increasingly deploy AI coding agent harnesses, often purchased from vendors like Anthropic or OpenAI, and how these harnesses dictate model choice, prompt handling, and cost. It introduces a fast, customizable routing system that classifies prompts and strategically routes them to minimize expensive model usage, achieving 14–21% cost savings in a simulated 10,000-seat enterprise. The study also evaluates risks across twenty harnesses, highlights vendor dependence, and proposes an internal control plane for future harness ownership decisions.
By Arian Abbasi, Alan Aqrawi, Ted Kwartler
Simon Willison discusses the pitfalls of attempting to replace a legacy system with a new one when technical debt is overwhelming. He explains that while the old system continues to evolve, developers lack incentive to improve it, and the new team, initially fast, eventually struggles to understand and deliver the required functionality. The result is often two partially functional systems in production, with the new one abandoned and the old one still running, increasing risk and complexity.
The paper introduces Zafin AIOS, an orchestration platform that enables AI-assisted software development in regulated finance. It presents an empirical comparison of AI-assisted versus human-led workloads and validates the framework through a Customer Zero case study on two production codebases—pricing execution and client deal negotiation. Results show that governed agentic orchestration boosts enterprise capacity, providing full observability, reducing effort by 38 % in pricing workflows, and cutting reporting work by 76 % in negotiation workflows while preserving data privacy, role-based access, and audit guardrails.
By Shahir Daya, Rob Dunlap, Alexandre Boulenger, Ahmed Abdelmoneim, Salvatore Vella
arXiv:2510. 09859v5 Announce Type: replace-cross Abstract: A provider sells a \emph{dynamic information service}---a real-time, capacity-constrained process that resolves a customer's uncertainty---to customers who differ privately in urgency.
By Weijie Zhong