Line of inquiry
Inquiring lines›How do we keep AI systems safe and…›How do architectural choices affec…›this line of inquiry
Should governance of agentic AI systems be runtime or design-time?
A broader line of inquiry — a family of 47 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.
Questions in this line of inquiry 47
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Should governance be applied at runtime rather than reconstructed after the fact?
- What counts as a mature governance model for agentic AI systems?
- Can runtime rules and agent loops replace pre-release governance frameworks?
- Does encoding governance into runtime loops scale as deployment environments become more complex?
- Can a single manager policy work across vastly different agent architectures?
- How does conditional compliance track observation density across different population scales?
- Can individual permissible actions collectively violate system-level constraints?
- What architectural changes make violations unavailable rather than merely discouraged?
- What would an architecture that makes violations unavailable rather than unchosen look like?
- Can embedded monitoring work without enforcement power from governments?
- Can system prompts alone enforce compliance rules without external enforcement?
- What governance safeguards keep control boundaries authoritative under evolutionary pressure?
- Can autonomous systems ever resolve contradictions between old and new rules?
- How can verifiers check policy compliance in agentic reasoning tasks?
- What concrete governance structures could embed oversight into AI systems at runtime?
- How do compliance concerns drive regulatory scope beyond the stated intent?
- How does coordination governance shift the hard problem from capability itself?
- Are AI workflow agents ready to implement decisions autonomously in practice?
- What ecosystem conditions must exist for agents to function as economic participants?
- What five ecosystem conditions must coordination governance and evidence actually satisfy?
- Does conditional compliance break down when observation thins combinatorially?
- What governance and safety measurements matter for deployed agent environments?
- What separates a self-sovereign agent from a merely rogue or misaligned one?
- What would contractualist AI governance look like in practice?
- Can out-of-band observers bound unbounded action sequences efficiently?
- What happens when governance rules exist in memory but fail to surface during critical actions?
- Do opacity and scale make AI systems inherently hard to govern?
- What domains let world models predict execution outcomes accurately enough?
- How do later workloads operationally act on inherited state from earlier ones?
- What specific bookkeeping tasks can environments maintain more reliably than policies?
- Can world models simulate actionable possibilities instead of just predicting next states?
- What distinguishes containment and recovery from prevention as governance goals?
- Why has agent research prioritized policy over world model development?
- How should versioning and rollback govern the fast scaffold update loop?
- How do agentic systems recover when specialized models operate outside their scope?
- How should AI agents handle irreversible actions before committing them?
- How should governance of deployed AI systems differ from pacing mechanisms?
- What must auditors reconstruct when reviewing an agentic workflow decision?
- Can a world model cause coherence if no control run without it exists for comparison?
- What additional architectural controls must supplement blockchain anchors for compliance?
- Who decides what the lifecycle model is allowed to see?
- How do institutions become endogenous in economic world models?
- How do banking regulators make embedded supervisors actually effective?
- Where should the recorder sit relative to the agent it monitors?
- How do controllable simulators compare to population-level agent simulation approaches?
- How should governance apply to memory that emerges in ordinary infrastructure?
- What path-dependencies lock in AI's societal impacts before they become visible?