A Permission Graph is a directed knowledge graph that formally represents discretionary rights or authorizations granted by law, where nodes are legal actors and directed edges are specific actions they are permitted to take. Unlike an Obligation Graph (which models mandatory duties) or a Prohibition Graph (which models forbidden acts), the Permission Graph exclusively encodes the deontic modality of permission, capturing the legal reality that an actor may—but is not required to—perform an action.
Glossary
Permission Graph

What is a Permission Graph?
A formal data structure for modeling discretionary legal authorizations within computational statutory interpretation systems.
Computationally, this structure enables normative reasoning engines to traverse statutory permissions and resolve complex regulatory queries, such as determining whether a specific entity has the authority to disclose protected data under a statutory safe harbor. The graph integrates with Deontic Logic formalisms and Rule-to-Fact Binding mechanisms to support automated compliance checking, where the absence of a permission edge can be as legally significant as the presence of an explicit prohibition.
Key Characteristics of Permission Graphs
A permission graph is a directed knowledge graph representing discretionary rights or authorizations granted by law, where nodes are actors and edges are actions they are permitted to take. The following characteristics define its computational architecture and distinguish it from obligation and prohibition graphs.
Deontic Modality: Permission
The foundational edge type in a permission graph is the deontic permission modality, formally distinct from obligation or prohibition. In deontic logic, permission is defined as the absence of a prohibition: an action P is permitted for actor A if and only if there is no norm stating that A must not do P. This is often modeled using the modal operator 'P'.
- Explicit Permission: A statute affirmatively grants the right (e.g., 'The Commissioner may issue guidance').
- Implicit Permission: An action is not addressed by any obligation or prohibition, making it legally permissible by default in most legal systems.
- Bilateral Permission: An actor is permitted to both perform and refrain from performing an action, representing true discretion.
Directed Edge Semantics
Every edge in a permission graph is a directed, labeled relationship from a source actor node to a target action node or another actor node. The directionality encodes the flow of authorization.
- Actor-to-Action:
[EPA Administrator] --(may issue)--> [Guidance Document]represents a direct grant of authority. - Actor-to-Actor:
[Principal] --(may delegate to)--> [Agent]models the transfer or sharing of discretionary power. - Conditional Edges: Edges can carry attributes representing preconditions that must be satisfied before the permission becomes active, such as 'upon a finding of good cause' or 'after public notice and comment.'
- Edge Weighting: In quantitative models, edges can be weighted to represent the scope of discretion, from narrow (specific enumerated actions) to broad (general regulatory authority).
Normative Conflict Resolution
Permission graphs must algorithmically resolve conflicts with other deontic modalities in the same legal corpus. A well-formed graph enforces deontic consistency rules to prevent logical contradictions.
- Permission vs. Obligation: If an action is both permitted and obligatory, the obligation takes precedence. The permission edge is retained but marked as superseded or redundant.
- Permission vs. Prohibition: A prohibition always defeats a permission. The graph must detect and flag normative conflicts where a permission edge and a prohibition edge connect the same actor to the same action.
- Lex Specialis Priority: Specific permissions override general ones. A statute granting a specific agency the right to waive a requirement takes precedence over a general prohibition on waivers.
- Temporal Precedence: Newer enactments override older ones. The graph must version edges by effective date to resolve temporal conflicts.
Actor Hierarchy and Inheritance
Permission graphs model organizational and legal hierarchies through inheritance of permissions. Subordinate actors may inherit the permissions of their principals, subject to constraints.
- Vertical Inheritance: A regional director may inherit the enforcement permissions of the agency head, unless explicitly limited by regulation.
- Delegation Chains: The graph traces multi-hop delegation paths:
[Secretary] --(may delegate to)--> [Deputy Secretary] --(may sub-delegate to)--> [Assistant Secretary]. - Scope Limitation: Inherited permissions can be narrowed but not expanded. A delegate cannot grant themselves more authority than the delegator possessed.
- Revocation Propagation: If a principal's permission is revoked, the graph must propagate the revocation to all downstream inherited permissions, requiring cascading invalidation logic.
Conditional Predicate Logic
Permissions are rarely absolute. The graph encodes conditional predicates as metadata on edges, representing the factual triggers that activate or deactivate a permission.
- Precondition Gates:
IF [emergency declaration is active] THEN [FEMA Director] --(may deploy resources)--> [Disaster Zone]. - Temporal Bounds: Permissions can have effective dates, sunset clauses, and renewal conditions that constrain their temporal validity.
- Jurisdictional Constraints: A permission may be limited to a specific geographic area, industry sector, or class of regulated entities.
- Procedural Prerequisites: Many permissions require procedural steps before exercise, such as notice-and-comment rulemaking, environmental impact assessment, or congressional notification.
Query and Traversal Patterns
Permission graphs support specific query patterns essential for computational legal reasoning and compliance checking.
- Permission Existence Query: 'Is actor A permitted to perform action X under statute S?' resolves to a boolean answer by checking for a valid, unconflicted permission edge.
- Full Permission Enumeration: 'What are all actions actor A is permitted to take?' returns the complete set of outgoing permission edges, including inherited permissions.
- Authority Gap Detection: 'What actions are neither permitted, prohibited, nor obligated for actor A?' identifies regulatory gaps where the law is silent.
- Delegation Path Tracing: 'Through what chain of authority did actor A receive permission P?' traverses the graph backward to find the originating statutory grant.
Permission Graph vs. Obligation Graph vs. Prohibition Graph
A structural comparison of the three core directed knowledge graphs used in computational deontic logic to model normative legal relationships.
| Feature | Permission Graph | Obligation Graph | Prohibition Graph |
|---|---|---|---|
Deontic Modality | Permission (May) | Obligation (Shall/Must) | Prohibition (Shall Not/May Not) |
Edge Semantics | Authorized Action | Mandatory Duty | Forbidden Action |
Formal Logic Operator | P (Permitted) | O (Obligatory) | F (Forbidden) |
Violation Consequence | No legal penalty for non-exercise | Legal penalty for non-performance | Legal penalty for performance |
Actor Agency | Discretionary | Compulsory | Restrictive |
Typical Statutory Trigger | Conditional grant of authority | Imposition of a duty | Criminal or civil liability clause |
Graph Traversal Purpose | Identify authorized actions | Identify required actions | Identify proscribed actions |
Complementary Relationship | F(p) ≡ ¬P(p) and O(p) ≡ ¬P(¬p) | O(p) ≡ F(¬p) | F(p) ≡ ¬P(p) |
Frequently Asked Questions
Clarifying the structure, function, and computational logic of permission graphs in statutory interpretation and regulatory compliance systems.
A permission graph is a directed knowledge graph that formally models discretionary rights or authorizations granted by law, where nodes represent legal actors (e.g., 'the Commissioner,' 'a licensee,' 'the Agency') and directed edges represent actions they are explicitly permitted to take under specific statutory conditions. Unlike an obligation graph, which encodes mandatory duties, a permission graph captures the permissive deontic modality—actions that are allowed but not required. Each edge is typically annotated with the statutory source citation, any precondition predicates that must be satisfied for the permission to be exercised, and the scope of the authorization. This structured representation enables automated compliance systems to answer queries like 'Is entity X allowed to perform action Y under regulation Z?' by traversing the graph from the actor node and checking for a valid permission edge with satisfied preconditions.
Enabling Efficiency, Speed & Accuracy
Intelligent Analysis, Decision & Execution
We build AI systems for teams that need search across company data, workflow automation across tools, or AI features inside products and internal software.
Talk to Us
Search across company data
Give teams answers from docs, tickets, runbooks, and product data with sources and permissions.
Useful when people spend too long searching or get different answers from different systems.

Automate internal workflows
Use AI to route work, draft outputs, trigger actions, and keep approvals and logs in place.
Useful when repetitive work moves across multiple tools and teams.

Add AI to products and internal tools
Build assistants, guided actions, or decision support into the software your team or customers already use.
Useful when AI needs to be part of the product, not a separate tool.
Related Terms
Understanding the Permission Graph requires familiarity with the broader deontic and structural concepts that govern computational legal reasoning. These related terms form the foundational vocabulary for modeling statutory authorization.

About the author
Prasad Kumkar
CEO & MD, Inference Systems
Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.
His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.
Partnered with leading AI, data, and software stack.
How We Work
Custom AI workflows for your Business
One-fit-all AI don't work for modern businesses. At Inferensys, we aim to understand your business & custom requirements; which we use to define most efficient agentic workflows, the data, and the tools for your business.
01
Review the use case
We understand the task, the users, and where AI can actually help.
Read more02
Pick the right approach
We define what needs search, automation, or product integration.
Read more03
Build the first useful version
We implement the part that proves the value first.
Read more04
Improve from there
We add the checks and visibility needed to keep it useful.
Read moreThe first call is a practical review of your use case and the right next step.
Talk to Us