The missing owner in agentic AI risk management

Q: What accountability gap does agentic AI create, and why doesn't a spending cap fix it?

The accountability gap is the missing owner of an agentic AI system's spending and stop rules. Most enterprises can name a budget holder, but not the person who decided what an autonomous system can spend, which decisions it can make without a human checkpoint, and when it must halt. That vacancy, not the invoice, is the real risk in agentic AI risk management, because individually reasonable choices aggregate into costs no procurement cycle tracked.

Q: What is the "integrator" role, and why is it suited to data scientists?

The integrator is a boundary-spanning role defined by organizational theorists in 1967 to resolve disagreement between divisions producing conflicting answers to the same question. Unlike a coordinator, translator, or orchestrator, the integrator sits inside the system and is accountable for outcomes. Data scientists already do this work: building evaluation harnesses, tracing false positives to corrupted data, and telling product teams a model isn't ready to ship.

Q: Why does the gap between benchmark performance and real-world reliability matter for regulated industries?

METR's 2026 research found that human maintainers evaluated AI-generated code on a time horizon roughly seven times shorter than automated benchmarks predicted. For regulated enterprises facing the EU AI Act's high-risk obligations and the FCA's live testing requirements, that gap doesn't close through model improvement alone. It closes when the institutional layer, escalation paths, stop rules, and audit trails, learns what real gatekeepers and regulators actually require before deployment.

Agentic AI can generate costs and decisions faster than any oversight process tracks, and most organizations practicing agentic AI risk management still cannot name who owns the resulting judgment calls or when a system must stop.

Eduardo Arino de la Rubia

Professor of Practice | Former Sr. Director of Data Science at Meta

Central European University

What you'll take away from this session

No one owns agentic AI's spending rules

The person who designs an agent's spending rules doesn't exist at most firms: they can name a budget holder, but not who decides what it can spend or when it must stop.

Undelegated judgment is the deeper risk

A spending cap fixes an invoice, but it leaves the underlying judgment gap in place: whether an agent's confidence was ever calibrated to the outcomes it affects.

Regulators are moving faster than programs

The FCA's live testing cohort and the EU AI Act's high-risk obligations converge on a 12- to 24-month window for firms to prove they can govern autonomous systems.

Technical fluency earns the integrator seat

Governance without technical depth is theater. Judging whether an agent's confidence is calibrated to real outcomes requires understanding how it fails, not just how it scores on a benchmark.

Benchmarks understate real-world failure rates

METR's 2026 research found maintainers judged AI code contributions on a time horizon roughly seven times shorter than automated benchmarks predicted.

Disagreement between agents is useful signal

Disagreement between agents trained on different data is signal, not noise: averaging it away discards the clearest evidence of where a decision genuinely needs human judgment.

Agentic AI risk management has a structural gap most organizations haven't filled: the person who designs the rules governing what an autonomous system can spend, which decisions it can make without a human checkpoint, and when it must stop. Eduardo Ariño de la Rubia argues the real risk isn't overspending, it's undelegation, handing decisions to systems before anyone has defined who owns the residual risk.

Spending caps fix the invoice; they leave that judgment gap in place, and no set of guardrails for agentic systems closes it by itself. A 2026 survey found most banking professionals cannot confirm their institution could shut down a malfunctioning model, and emerging model risk management guidance is starting to force the question.

Ariño makes the case that data scientists are the natural occupants of a role organizational theorists named in 1967: the integrator, accountable for outcomes rather than communication. The session closes with six diagnostic questions for calibrated confidence before any agent acts on a consequential decision.

FAQ

What accountability gap does agentic AI create, and why doesn't a spending cap fix it?

What is the "integrator" role, and why is it suited to data scientists?

Why does the gap between benchmark performance and real-world reliability matter for regulated industries?

Transform the work that matters most

See how Domino helps the world’s most regulated enterprises build, scale, and govern AI-powered applications.

Watch demo Talk to an expert