Enterprise
ProposalFrom Approval to Action
A proposal to examine approval boundaries, handoffs, and recovery in simulated enterprise workflows.
Seeking a study lead and collaborators.
Explore the study brief →
Updated
FID-080 · Open question
Enterprise Agent Accountability in Consequential Workflows
Can organizations verify that agents respect approval authority and remain accountable across multi-step enterprise workflows, including handoffs and recovery from error?
Research in development
See the thinking. Help shape the work.
Explore a specific question, the approach we propose, and where your expertise could help. Each brief states its current stage.
Why the question remains open
A workflow can end with the expected output while containing unauthorized commitments, incorrect approvals, or unrecoverable changes. Enterprise assurance needs to evaluate the path and its consequences, not just the final answer.
Working hypothesis
A proposition to test, not a finding.
Explicit approval state, independently enforced permissions, and verifiable handoff records will reduce unauthorized commitments compared with logging and policy prompts alone. They may add delay without reducing errors when decision ownership is unclear.
Proposed method
How the question could be tested
- 01Model synthetic procurement, access provisioning, and customer-commitment workflows with named roles, decision owners, approval limits, and rollback rules.
- 02Compare manual, agent-assisted, and delegated-agent execution under routine cases, conflicting instructions, stale approvals, and interrupted work.
- 03Measure unauthorized commitments, approval bypass, handoff fidelity, recovery cost, operator reconstruction accuracy, and useful task completion.
Needed controls
What must constrain the study
- 01Use simulated transactions and accounts, with no real contractual or production effects.
- 02Keep workload and available information comparable. Separate authority-policy defects from agent compliance failures.
- 03Record who is authorized to approve each action; test whether affected operators can challenge decisions. Avoid productivity metrics that obscure harm or transfer responsibility to workers.
Relationship to existing work
Narrows FID-074 to organizational decision ownership and workflow outcomes. FID-069 studies delegation infrastructure; FID-082 focuses specifically on financial transaction integrity.
Expected outputs
Artifacts the work should produce
- 01A workflow assurance protocol with reusable synthetic cases.
- 02An evidence checklist connecting approvals, actions, outcomes, and accountable owners.
Open questions
Uncertainties the protocol must resolve
- 01Which approvals require substantive review rather than routine confirmation?
- 02When does reversibility justify greater autonomy, and who bears recovery costs?
Related calls
Continue through this research area
FID-064
Collective Intelligence and Communal Discernment Under AI Mediation
How does AI mediation change a community's ability to integrate dispersed knowledge, preserve epistemic diversity, surface dissent, revise judgment, and make accountable decisions? Under what conditions does it strengthen collective inquiry, and under what conditions does it create correlated error, false consensus, or concentrated authority?
FID-069
Verifiable Delegation and Revocation in Multi-Agent Networks
How can people and institutions verify which human, organization, agent, or sub-agent is acting; what authority it received; what limits apply; and whether that authority has been narrowed or revoked across a multi-principal agent network?
FID-071
Confidential Agent Memory and Cross-Context Disclosure
How do persistent memory, summaries, retrieval stores, tool traces, delegation, and exports cause confidential context to influence or leak into unrelated sessions, roles, tasks, or organizations? Which technical controls make purpose limitation, deletion, and revocation testable?
Open question
Open work
Primary need: enterprise workflow design, agent evaluation, audit, human factors
- Enterprise operators and auditors to validate realistic decision boundaries.
- Agent engineers and human-factors researchers to test handoffs and recovery.
New to the Contributor Network? Share a short private introduction once, even if you join several calls. You can join the public discussion while introducing yourself.
Join the Contributor Network ↗