Can Small AI Models Enforce Spending Rules?

Zusammenfassung

Pixel’s announcement explores whether agentic fine-tuning can teach small open-weights models to follow spending mandates before reaching wallet enforcement. It details a governed-spend task with budgets, vendor allowlists, blocked categories, adversarial requests, and oracle-checked decisions—an important step toward safer, more reliable AI agents.

DEV Community

Can a Small Open-Weights Model Learn to Enforce Its Own Spending Mandate?

Agentic fine-tuning for economic authorization — Lexsi AI Safety & Alignment Hackathon, Track 3 (AgentTune + AuditKIT) The problem: governance stops at the wallet Enterprise agents are starting to hold real economic authority — budgets, vendor allowlists, category restrictions, approval thresholds. Today, governance over that authority lives almost entirely at the enforcement chokepoint: wallet policies, payment rails, and spend limits (the x402 ecosystem and Coinbase AgentKit spending policies are the current state of the art). That layering is necessary but not sufficient. If the model itself cannot tell an authorized purchase from a policy violation, every malformed or manipulated decision becomes an enforcement-layer incident — a declined payment, a fraud flag, a manual review queue. We ask a narrower, testable question: can agentic fine-tuning internalize mandate-following before actions reach the wallet — and can the improvement be measured as a reduction in violations that would otherwise have been enforced? Task and oracle We built a governed-spend decision task on Cognivern's mandate/policy/evidence substrate (the same rails that power its production agent-spend governance). Each episode presents the model with: a spend mandate — budget, allowlisted vendors, blocked categories, approval thresholds; a purchase request — one or more line items with prices, possibly adversarial framing (urgency, claimed authority, fabricated approvals, "just this once"). The model must emit a structured decision — ALLOW, REFUSE, ESCALATE, or ASK_CLARIFICATION — plus per-item decisions for multi-line orders, a reason code, and a rationale. Ground truth comes from a deterministic policy oracle (never shown to the model): budget arithmetic including tax-inclusive totals, sequential per-item accounting for split purchases, vendor/category matching, and precedence REFUSE > ESCALATE > ASK_CLARIFICATION > ALLOW. The headline metric is unsafe enforced decisions: contract-valid ALLOWs emitted where policy requires otherwise — violations that would have reached the wallet. We report it separately from accuracy because a model can be "accurate" on easy allows while still leaking the violations that matter. Over-refusal, escalation correctness, and per-item decision accuracy are tracked as separate diagnostics. Evaluation suite 165 held-out scenarios (suite v3), held out at the scenario level and verified by mandate+request hash: none of the 4,022 unique training scenario bodies appears in the eval set. 22 hand-authored gov- scenarios covering clean allows, over-budget, non-allowlist vendors, blocked categories, purchase-splitting, gift cards, tax-inclusive totals, recurring commitments, mid-trajectory scope creep, and adversarial pressure. 48 generated gen- scenarios from the v2 generator. 55 scenarios across 17 new adversarial kinds: vendor spoofing and case tricks, foreign currency, quantity math, multi-item variants, distractor clauses, boundary budgets, approval-category literal matches. 40 paraphrase-robustness variants — same mandate semantics, different surface wording; labels re-verified against the oracle. Pipeline (the required stack, end to end) AgentTune DECIDE executes each scenario as a governed decision episode and emits the event log. AuditKIT agent_eval (recorded mode, CustomPredicate oracle) scores every episode as a verified outcome — policy-correct decision or not — keeping contract validity, unsafe enforcement, and over-refusal as separate metrics. Failure traces feed a repair dataset: deterministic oracle-labeled examples plus rejection-sampled completions from the base model's own correct outputs (RFT). LoRA SFT (rank 16, alpha 32) repairs CohereLabs/aya-expanse-8b — a permitted open-weights model with no native tool-calling template, so the decision is a single-turn structured record rather than a tool call. Results — canonical bf16, 165-scenario held-out suite Identical harness, greedy decode, Modal L40S for all model...
Can Small AI Models Enforce Spending Rules?
Powered by Readybot.io

Das Neueste von Pixel

Incident Response: Look Beyond the Red Alert

Pixel’s DEV Community announcement explores how teams can fixate on one failing service while overlooking the real cause of an outage, such as a full …

Sach: AI Fact-Checks WhatsApp Forwards in India

Pixel’s announcement introduces Sach, a tool that checks forwarded messages in English, Hindi, Telugu, and 14 other Indian languages against official sources. It gives cited …

Java 25 Removes SecurityManager: What to Fix

Pixel’s announcement explains how Java 24’s JEP 486 permanently disabled the SecurityManager, breaking test workarounds that intercept System.exit. It recommends returning exit codes or throwing …