Beyond Training: A Feasibility Taxonomy for Inference-Time AI Governance
A study maps twenty inference-time AI governance mechanisms, finding commercial readiness only against cooperative deployers and no adequate defense versus state-level adversaries.
The paper develops a feasibility taxonomy of twenty inference-time AI governance mechanisms across monitoring, verification, and enforcement, each rated on a four-point readiness scale against a four-vendor evidence base. Fifteen of the twenty mechanisms have commercial technical substrates in production today, though governance-grade assurance and adversarial robustness vary substantially. Stress testing shows readiness holds only against a cooperative deployer and low-to-medium-capability user: no mechanism rates adequate against a high-capability state-level deployer, and fine-tuning removes model-internal enforcement components. A second-rater reliability check on readiness ratings returned a quadratic-weighted Cohen's kappa of 0.74.
- Twenty inference-time governance mechanisms rated for readiness across four governance scenarios
- Fifteen mechanisms already have commercial technical substrates in production
- No mechanism rated adequate against high-capability state-level deployers
- Fine-tuning bypasses model-internal enforcement, though platform-external controls can persist
From the original article
1 min read at arxiv.orgCompute governance today is a governance of training: the thresholds, reporting requirements, and frontier-AI regimes now in force attach to training compute and treat the trained model as the regulatory unit. That picture is incomplete: capability increasingly migrates to the deployment stage through inference-time scaling, agentic scaffolding, and compression onto consumer hardware.
By Samar Ansari · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.