Claim charts
Element-by-element claim coverage for 8 nonprovisional applications and 3 active provisionals. Each independent claim broken into its constituent limitations and mapped to industry convergence.
Published
Published applications
19/326,036 — Adaptive Network Framework For Modular, Dynamic, and Decentralized Systems
Claim 1 of 18 (Method) — Mutation governance in decentralized systems
| Element | Convergence |
|---|---|
| Registering an anchor object to a container within an adaptive index configured to validate alias mutations and resolve identifier collisions | DNS validates name records but has no concept of mutation governance or collision resolution under quorum. This is a governed naming system where structural changes require consensus. |
| Adaptive index with entries organized in parent-child hierarchy, each corresponding to a unique semantic scope identified by a structured alias | Kubernetes namespaces, AWS organizations — all use flat or shallow hierarchies. Deeply nested semantic scopes with structured aliases are novel and required for fine-grained multi-tenant agent environments. |
| Evaluating mutation proposal according to policy with quorum validation procedures | Blockchain consensus validates transactions. This validates structural mutations (reorganizations, splits, merges) under quorum. No existing system governs structural change to the index itself by consensus. |
| Performing structural mutation (segmentation, merging, or relocation) while preserving lineage continuity | Database sharding splits data but loses structural lineage. This preserves the full history of how containers were split, merged, or relocated — critical for auditability in regulated environments. |
Claim 8 of 18 (Platform) — Adaptive network platform
| Element | Convergence |
|---|---|
| Semantic indexing module with adaptive index organizing assets into nested containers by structured alias and semantic scope; hierarchical namespace with dynamic reclassification | LDAP, Active Directory organize by administrative structure. This organizes by semantic meaning with dynamic reclassification under governance — the namespace restructures itself based on what the assets actually are. |
| Mutation governance module evaluating structural changes based on quorum thresholds and lineage consistency | No existing platform validates structural reorganization against lineage consistency. Every enterprise system that must restructure without breaking audit trails converges here. |
| Telemetry orchestration module triggering routing adjustments and cache instantiation based on mutation rejection rates, response latency, storage utilization | Prometheus/Grafana monitor but don't actuate. This closes the loop — telemetry directly triggers routing and caching decisions under policy. Required for self-healing distributed agent networks. |
| Operating without centralized control; continuous reconfiguration based on demand, proximity, and anchor-local governance | Fully decentralized operation with local governance. Kubernetes requires a control plane. This does not. |
Claim 15 of 18 (System/CRM) — Adaptive network system
CRM claim covering: symbolic alias registration, structural mutations via quorum, dynamic query routing, device/agent attribute identification, decentralized access controls, ephemeral cryptographic hash authentication. 13 dependent claims cover delegation chains, trust-scoped routing, mutation rollback, anchor migration, and cross-zone federation.
19/366,760 — Cognition-Compatible Network Substrate and Memory-Native Protocol Stack
Claim 1 of 20 (System) — Memory-native protocol execution
| Element | Convergence |
|---|---|
| Plurality of agents each with unique identifier, payload, memory field, transport header, and cryptographic signature | HTTP messages have headers and payloads but no memory, no identity, no signature. These are agents-as-protocol-units: every message carries its own identity, history, and authentication. |
| Modular protocol stack (routing layer, indexing layer, consensus layer) executed at each distributed node | TCP/IP separates transport from application. This separates routing, indexing, and consensus as modular layers — each independently governable. The protocol stack for agent-native networks. |
| Behavior determined by metadata embedded within received agents | In HTTP, the server determines behavior. Here, the agent itself carries instructions that govern how it is routed, indexed, and validated. The network serves the agent, not the other way around. |
| Memory field includes verifiable lineage, access logs, and policy references governing routing, mutation, and consensus | No existing protocol embeds governance into the transport. Policy-at-the-protocol-level means every hop, every mutation, every consensus decision is governed by what the agent carries. |
Claim 14 of 20 (Method) — Distributed memory-native communication
| Element | Convergence |
|---|---|
| Receiving agent at node; verifying signature; parsing transport header and memory field | TLS verifies the connection. This verifies the agent itself — signature validation is per-agent, not per-channel. |
| Determining routing eligibility and mutation scope by evaluating access log and policy references | IP routing uses destination addresses. This routes by evaluating the agent's history and policy. Where the agent can go depends on where it has been and what its policy permits. |
| Forwarding to eligible nodes determined by dynamic routing protocol and memory field constraints | The agent's own memory constrains its routing. No existing protocol allows the message itself to restrict its own forwarding path based on its accumulated history. |
18 dependent claims cover trust-zone scoping, lineage verification, consensus participation, cache governance, cross-zone agent migration, and policy-constrained multicast.
19/388,580 — Systems and Methods for Memory-Native Identity and Authentication
Claim 1 of 21 (Method) — Memory-native two-stage authentication
| Element | Convergence |
|---|---|
| Generating DAH_t as successor of prior DAH_{t-1} under update rule incorporating unpredictability and volatile salt | Passwords are static. TOTP rotates on a shared secret. DAH evolves from its own trajectory — each hash is a successor of the last, incorporating environmental unpredictability. No shared secret, no rotation ceremony. |
| Deriving symmetric key from recipient's current dynamic identity (DAH_R or DDH_R) | TLS negotiates keys via handshake. Signal uses Diffie-Hellman. This derives the key from the relationship's dynamic identity state — no negotiation needed because both sides can compute it independently. |
| Message does not include the symmetric key; recipient reconstructs from locally retained trust-slope state | Every modern crypto protocol transmits or negotiates the key. This never transmits it. Interception yields ciphertext with no path to the key. |
| Validating DAH_t against expected successor candidate within policy-bounded continuity parameters | Certificate pinning validates a static identity. This validates a trajectory — the credential must be a plausible continuation of the sender's history. Stolen credentials from the past don't produce valid futures. |
| Extracting embedded DAH_S from decrypted payload and validating against reconstructed trust slope | The second stage: even after decryption, the embedded identity must independently validate against the trust slope. Two independent verification paths, both trajectory-based. |
Claim 16 of 21 (System) — Agent mutation entanglement
| Element | Convergence |
|---|---|
| Host device computing DDH_t as successor of DDH_p incorporating unpredictability and volatile salt | Device identity today is a static certificate or TPM key. This makes device identity dynamic — the device evolves its identity hash at every mutation, incorporating local entropy. |
| Semantic agent computing DAH_s from prior DAH_p and host mutation token derived from DDH_t | Agent and device identities become cryptographically entangled. The agent's identity evolution depends on the device's identity evolution. You cannot separate the agent from the device it runs on without breaking the entanglement. |
| Entanglement trace recording DAH_p, DDH_t, host mutation token, and mutation metadata; validator accepting DAH_s only if trace opens to DDH_t under policy | Hardware-anchored agent identity. No existing system creates a cryptographic proof that an agent's identity evolved on a specific device. Required for any agent trusted to act autonomously on hardware it doesn't own. |
19 dependent claims cover multi-device trust-slope synchronization, trust-slope recovery after device loss, credential-less roaming, and policy-governed identity migration.
Filed, Awaiting Publication
Pending applications
19/230,933 — Cognition-Native Semantic Execution Platform for Distributed, Stateful, and Ethically-Constrained Agent Systems
Claim 1 of 15 (Method) — Executing cognition-native semantic agents
| Element | Convergence |
|---|---|
| Instantiating a semantic agent with intent field, context block, memory field, policy reference field, mutation descriptor field, and lineage field | The foundational six-field agent definition. No existing agent framework defines agents as structured objects with canonical fields that carry their own governance, history, and mutation rules. |
| Evaluating policy reference at runtime prior to any mutation, delegation, or propagation; deterministic permit/deny | Policy-first execution. LangChain evaluates output after generation. This gates every action before it happens. Nothing executes without policy clearance. |
| Resolving human-readable aliases via adaptive indexes with anchor-based consensus, slope-indexed pathfinding, and entropy-derived identifier registration | The semantic DNS for agent systems. Aliases resolved by meaning within governed scopes using trust-slope-based pathfinding. Required for any system where agents discover each other by purpose. |
| Routing across memory-native substrate based on semantic trust scope, contextual relevance, and mutation eligibility | HTTP routing uses URLs. gRPC uses service names. This routes by trust scope and mutation eligibility — the agent goes where its policy permits and its context is relevant. |
| Validating semantic lineage through entropy-resolved trust slope verification | Identity by trajectory, not by credential. The agent proves its authenticity through the continuity of its behavioral history, not through a static token. |
Claim 9 of 15 (Platform) — Cognition-native semantic execution platform
| Element | Convergence |
|---|---|
| Memory-bearing semantic agents with six canonical fields | The agent object definition. Every agent in the platform is a structured, self-describing, policy-carrying object — not a prompt-response pair. |
| Memory-native substrate routing and storing agents across distributed trust zones | The network layer. Agents are first-class network citizens — routed, stored, and governed by the substrate based on their own fields. |
| Governance layer with cryptographically signed policy objects defining mutation permissions and semantic constraints within scoped trust zones | The governance layer. Signed policies, not prompt instructions. Scoped to trust zones, not globally. Cryptographically verified, not trusting the caller. |
| Distributed indexing layer with adaptive indexes mapping aliases to identifiers, governed by entropy-sensitive anchors validating alias mutations and enforcing lineage policies | The indexing layer. Adaptive indexes that validate their own mutations under anchor consensus — a self-governing naming system. |
| Entropy-resolved identity layer authenticating lineage via behavioral trust slopes without persistent static credentials | The identity layer. No passwords, no certificates, no API keys. Identity is the trajectory. |
13 dependent claims cover execution failure recovery, cross-zone delegation, adaptive caching, telemetry-driven routing, and mutation rollback.
19/452,651 — Cognition-Compatible Semantic Agent Objects with Structural Validation, Partial Agent Support, and Traceable Semantic Lineage
Claim 1 of 26 (System) — Cognition-compatible semantic agent objects
| Element | Convergence |
|---|---|
| Semantic agent object with embedded canonical fields (intent, context, memory, policy reference, mutation descriptor, lineage) | JSON Schema, Protobuf require external schemas. This agent object is self-describing — structural validity determined from information embedded within the object itself. |
| Node determining structural coherence based on field presence and structural compatibility; determined based only on information within the agent object | No external schema registry. No version negotiation. The object carries everything needed to validate itself. Required for agents operating in unknown or heterogeneous environments. |
Claim 19 of 26 (Method) — Validating semantic agent objects
| Element | Convergence |
|---|---|
| Determining structural validity from presence and coherence of canonical fields | Structural validation without prescribed execution order. The object is valid or invalid based on its fields, independent of runtime context. |
| Determining mutation eligibility using policy reference and mutation descriptor fields | The object enforces its own mutation policy. Not "the orchestrator decides" but "the object knows what changes it permits." |
| Recording outcomes in memory field; performed without prescribing execution order, scheduling, or runtime control | Execution-order-independent validation. The agent object can be validated at any node, in any order, without coordination. Required for truly distributed agent systems. |
Claim 18 (CRM). 23 dependent claims cover partial agent support, traceable semantic lineage, field-level governance, and cross-platform portability.
19/538,221 — Memory-Resident Execution of Persistent Executable Objects In Distributed Computing Systems
Claim 1 of 20 (Method) — Semantic object-resident computation
| Element | Convergence |
|---|---|
| Persistent executable object with intent field encoding machine-readable execution descriptor, context block, and memory field encoding prior execution state | Lambda functions are stateless. Docker containers mount external state. This object carries its own execution state internally — prior execution records travel with the object. |
| Propagating among plurality of execution nodes | Mobile agents (1990s research) moved code. This moves code, state, history, policy, and identity as a single object. The agent is not deployed to a node; it arrives with everything it needs. |
| Execution evaluation cycle: parse intent, evaluate context against local policy, read memory for prior records, select action from {execution, mutation, delegation, dormancy, reentry, termination} | The six-action execution model. Execute, mutate, delegate, go dormant, re-enter, or terminate. Current frameworks offer execute-or-fail. Dormancy and reentry as first-class outcomes are required for long-lived agents. |
| Execution continuity maintained by the memory field of the persistent executable object | Object-resident continuity. The agent carries its own history — no external database, no orchestrator state, no cloud dependency. Continuity is a property of the object, not of the infrastructure. |
Claim 20 of 20 (System) — Distributed computing system with persistent execution objects
| Element | Convergence |
|---|---|
| Persistent execution object with intent field, context block, and append-only memory field propagating among execution nodes | System claim covering the same architecture from the infrastructure side. The execution nodes serve the object; the object does not serve the nodes. |
| Each node: parse intent, evaluate context, read memory, select action, append outcome — without centralized coordination | No orchestrator. No control plane. Each node evaluates independently based on what the object carries. Fully decentralized execution. |
Claim 19 (CRM). 17 dependent claims cover dormancy triggers, reentry conditions, delegation protocols, cross-node lineage verification, and execution-failure recovery.
19/561,229 — Cryptographically Enforced Governance for Autonomous Agents And Distributed Execution Environments
Claim 1 of 20 (Method) — Cryptographically enforced governance
| Element | Convergence |
|---|---|
| Receiving proposed action associated with agent object; policy references including canonical aliases | Governance starts with the action and the policy that governs it. The policy is referenced by alias, not hardcoded — enabling dynamic, updatable governance. |
| Resolving policy references to obtain candidate external policy objects | Policy is external and resolved at runtime. Not baked into the model. Not in a prompt. A separate, signed object that can be updated independently of the agent or the model. |
| Filtering by freshness (validity window, revocation state, anti-rollback monotonicity) | Expired certificates block TLS. This applies the same freshness logic to governance policy — stale policy, revoked policy, or rolled-back policy is rejected before the action is even considered. |
| Cryptographic verification of policy authenticity | The policy must be signed by an authorized issuer. You cannot inject a fake policy. No prompt injection, no jailbreak, no social engineering can override cryptographically verified policy. |
| Determining authorization prior to enabling performance, prior to instantiating execution context or capability context | The governance gate. The agent cannot even begin to act until policy clears. Not "act and check" but "check then act." Required for any system where unauthorized actions have consequences. |
| Denial as a valid non-execution outcome | "No" is a valid answer. Current AI systems treat refusal as an error. This makes non-execution a first-class system outcome — architecturally, not as a hack. |
Claim 8 of 20 (System) — Cryptographically enforced governance for semantic agents
| Element | Convergence |
|---|---|
| Semantic agent object with intent, memory, lineage, and policy fields; policy comprising canonical aliases | The governed agent. Policy is a structural field of the agent, not an afterthought. The agent carries its own governance references. |
| Policy resolution component; verification component; governance gate | Three components: resolve, verify, gate. Separation of concerns — resolution is distinct from verification is distinct from authorization. No single point of bypass. |
| Governance gate deterministically permits or denies prior to instantiating execution or capability context; denial results in non-execution | The structural enforcement point. Every action (execution, mutation, delegation, propagation) must pass the gate. The gate is deterministic — same inputs always produce same authorization decision. |
Claim 6 (CRM). 17 dependent claims cover policy delegation chains, temporal policy constraints, multi-policy composition, escalation protocols, and governance audit trails.
19/647,395 — Systems and Methods for Autonomous Agents with Persistent Cognitive State, Self-Regulated Execution, and Cross-Domain Behavioral Coherence
The broadest filing in the portfolio — 15 chapters, 3 independent claims covering autonomous agents with persistent cognitive state and self-regulated execution. Also filed as PCT/US26/22839.
Claim 1 of 20 (System) — Autonomous agents with persistent cognitive state
| Element | Convergence |
|---|---|
| Cross-domain coherence engine with defined coupling functions maintaining bidirectional feedback between persistent cognitive domain fields | No existing agent models cognition as coupled domains with bidirectional feedback. GPT, Claude, Gemini have no persistent internal state, let alone coupled domains that influence each other. |
| Cognitive domain fields encoding behavioral disposition, normative alignment, and execution readiness as continuously updated persistent state | Three dimensions of cognition: how the agent behaves (disposition), what it considers right (normative), and whether it's ready to act (readiness). All persistent. All evolving. No current system tracks any of these. |
| Execution substrate validates proposed state transitions without retaining authority over agent's cognitive state | The substrate serves the agent, not the other way around. The agent owns its cognitive state; the substrate validates transitions. Inversion of the current model where the platform controls the agent. |
| Composite admissibility evaluator integrating signals from coupled domains to permit, gate, or suspend mutations | Current AI safety: "don't do X" in the prompt. Composite admissibility: the agent's own cognitive state across all coupled domains determines what it can do. Safety as a structural property. |
| Non-executing cognitive mode for speculative evaluation (branching hypotheticals, structured inquiries, delegation alternatives) without committing state changes | "Think without acting." No current AI system has a governed mode where the agent reasons about actions it is not confident enough to take. Required for consequential-action domains. |
| Every mutation, admissibility determination, and domain update recorded in lineage for deterministic behavioral reconstruction | Complete behavioral audit trail. The agent's entire cognitive history is reconstructible from lineage. Required for regulated industries, legal accountability, and forensic analysis. |
Claim 2 of 20 (Method) — Governing execution through cross-domain coherence
| Element | Convergence |
|---|---|
| Maintaining agent with persistent state, lineage, and cognitive domain fields coupled through bidirectional feedback; migratable between execution substrates while preserving behavioral continuity | Agent portability with behavioral continuity. The agent moves between devices/platforms and stays the same agent. No existing system guarantees behavioral continuity across migration. |
| Receiving proposed mutation; propagating through coherence engine for independent contribution from each domain; responsive updates between domains | Every mutation is evaluated holistically. Affect, integrity, confidence, capability all weigh in through feedback. Not a checklist — coupled domains that influence each other's assessment. |
| When suspended: iteratively generating candidate alternatives via branching hypothetical sequences, evaluating through composite admissibility, repeating until satisfying candidate found or external intervention | Self-correcting speculative planning. The agent doesn't just refuse — it generates alternatives until it finds one that satisfies all its cognitive constraints. Or it asks for help. No current system does this. |
Claim 3 of 20 (CRM) — Three-phase corrective loop
| Element | Convergence |
|---|---|
| Detecting deviation where cognitive domain field deviates from normative alignment defined by policy constraints | Drift detection. The agent monitors itself for normative drift. No current AI system has self-monitoring for deviation from its own values. |
| Three-phase corrective loop: detection phase, recording phase committing deviation as immutable truth, corrective pressure phase producing candidate mutation evaluated against composite admissibility | Honest self-correction. Phase 2 is critical: the deviation is recorded as truth before correction. The agent cannot pretend it didn't drift. Then it applies corrective pressure governed by admissibility. No existing system commits errors to permanent record before attempting correction. |
| Corrective pressure propagating through bidirectional feedback to modulate behavioral disposition toward restored normative alignment | Cross-domain correction. A deviation in one domain propagates corrective pressure through all coupled domains. The agent corrects holistically, not in isolation. |
17 dependent claims cover confidence thresholds, capability/time/uncertainty awareness, affect modulation, LLM skill gating, inference-time execution control, biological identity via trust-slope, semantic discovery, training governance, and disruption modeling.
Chapter coverage
The 15 chapters teach the following primitives, each supported by the independent claims above and extended by dependent claims:
Ch. 1 Foundation — Cross-domain coherence engine, composite admissibility, persistent cognitive domain fields
Ch. 2 Affect — Emotional modulation as persistent structural state, slope-validated affect transitions
Ch. 3 Integrity — Empathy-weighted integrity modeling, deterministic moral trajectory forecasting, three-phase corrective loop
Ch. 4 Forecasting — Branching hypothetical state sequences evaluated through coupled domains
Ch. 5 Confidence — Non-executing cognitive mode, confidence-governed dispatch, structured abstention
Ch. 6 Capability & Time — Resource-aware, time-aware, uncertainty-aware execution planning
Ch. 7 LLM Skill Gating — LLMs as governed tools gated by the agent's policy and cognitive state
Ch. 8 Inference Control — Semantic execution control at inference time, governing the forward pass itself
Ch. 9 Biological Identity — Continuity-based human identity using trust-slope validation
Ch. 10 Discovery — Semantic agent discovery by capability within governed trust scopes
Ch. 11 Training Governance — Depth-selective training governance under semantic policy
Ch. 12 Disruption Modeling — Predictive cascading failure modeling across agent networks
Ch. 13 Applications — Domain-specific embodiments (medical, legal, financial, industrial, vehicular)
Ch. 14 Platform Synthesis — Full composition of all primitives with cross-domain coherence
PCT/US26/28630 — Structural Content Identity and Rights-Grade Admissibility for Digital Artifacts
Claim 1 of 20 (System) — Computable content identity
| Element | Convergence |
|---|---|
| Content encoder deriving unique identifier from multi-axis variance vector (energy distribution, frequency compaction, structural phase persistence) as deterministic slope-indexed fingerprint | Perceptual hashing (pHash, SSCD) captures similarity but produces fixed binary digests. This fingerprint is a continuous variance vector enabling direct cosine similarity — both an identifier and a coordinate in content space. |
| Fingerprint enabling direct cosine similarity without decoding a fixed binary digest; operable on raster images, audio, text, video, and binary objects | Cross-modal content identity. One fingerprinting system across all media types. No existing system provides a unified, similarity-comparable identifier across modalities. |
| Anchor nodes scoped to variance bands within global slope continuum; cache memory storing lineage graphs, alias registrations, and policy constraints | Decentralized content governance. Anchors manage content within their variance band — a self-organizing registry where content finds its own neighborhood. |
| Alias resolution engine mapping human-readable identifiers to variance-derived UIDs under signed policy enforced by anchor quorum | Content naming under governance. You can name content, but naming is governed by quorum consensus and cryptographic policy. No unilateral renaming, no alias hijacking. |
| Provenance validator constructing multi-root lineage graphs via slope vector proximity and mutation deltas | Content provenance as a graph, not a chain. Multiple roots — content can derive from multiple sources simultaneously. The lineage graph tracks how content evolved, split, merged, and transformed. |
Claim 2 of 20 (Method) — Computable content identity method
| Element | Convergence |
|---|---|
| Normalizing content to canonical scalar field (modality-specific: luma-weighted grayscale for images, spectrogram for audio, token frequency grid for text, temporal delta for video) | Each modality gets a specific normalization that reduces it to a scalar field. This makes all content comparable regardless of format — the scalar field is the common representation. |
| Extracting multi-axis variance vector: energy distribution, frequency compaction, structural phase persistence across nested grid resolutions | Three axes of structural analysis at multiple scales. Not pixel-level comparison but structural behavior across scales. Survives cropping, resizing, transcoding, format conversion. |
| Deriving UID via weighted triads and multi-scale quantization hashing; assigning to slope band; registering with anchor nodes | The fingerprint becomes a position in a governed content space. Registration with anchor nodes means the content is now discoverable, traceable, and governable within its variance neighborhood. |
| Constructing multi-root lineage graph via cosine similarity above semantic continuity threshold | Automatic provenance discovery. If two pieces of content are structurally similar enough, they are linked in the lineage graph. No manual annotation required. |
| Enforcement through anchor quorum consensus of cache, alias, delegation, and temporal validity constraints; without centralized registries or static credentials | Fully decentralized content governance. No central registry, no static credentials. Everything governed by quorum and policy at the anchor level. |
Claim 3 of 20 (Platform) — Rights-grade content admissibility
| Element | Convergence |
|---|---|
| Pre-release admissibility engine evaluating candidate artifacts against signed policy objects prior to external commitment | Content gatekeeping. Nothing goes out until policy clears. Admissible categories, restricted classes, jurisdictional constraints, similarity thresholds, override authorities — all defined in signed policy. |
| Structural similarity evaluator computing similarity via variance-derived UIDs; rejecting, regenerating, or escalating if similarity exceeds threshold | Plagiarism detection as a structural operation. Not fuzzy text matching but variance-vector cosine similarity. Reject, regenerate, or escalate — three governed responses to excess similarity. |
| Training corpus governance layer admitting artifacts under signed corpus policies with cryptographically verifiable lineage | The AI training data defense. Every artifact in a training corpus must be admitted under signed policy with verifiable provenance. The legal shield for defensible training practices. |
| Consultation event logger recording variance-derived UID of consulted artifact, authorizing policy object ID, and timestamp for each generation event | Generation audit trail. When the model consults a reference during generation, the consultation is logged with the specific artifact, the specific policy that authorized it, and the timestamp. No existing AI system does this. |
17 dependent claims cover temporal validity, jurisdiction-specific policy, multi-modal normalization variants, lineage graph traversal, and corpus policy versioning.
Provisional
Active provisionals
Priority-establishing filings. Claim scope to be defined in nonprovisional conversion.
64/049,409 — Governed Spatial Mesh for Physical-World Perception, Coordination, and Actuation
Key disclosed elements
| Disclosed Element | Convergence |
|---|---|
| Three-tier governed spatial navigation: perception, coordination, actuation | ROS separates perception and actuation but has no governance layer. This adds governed coordination — what the robot perceives, how it coordinates, and what it actuates are all policy-governed. |
| Dispositional field derived from accumulated sensor lineage | Autonomous systems act on sensor snapshots. This derives a persistent spatial disposition from lineage — the agent's relationship to space evolves over time. |
| Fleet-contributed skill emergence from distributed observation | Skills emerge from fleet-wide observation, not central training. One robot's experience becomes the fleet's capability under governance. |
| Mesh-derived spatial coordinates independent of GPS | Coordinates derived from the mesh topology itself. GPS-denied environments (indoor, underground, space) need a coordinate system that comes from the agents, not from satellites. |
| Operator incapacitation detection; governed actuation under integrity constraints | The spatial mesh detects when a human operator is incapacitated and governs what the system does next. Required for safety-critical autonomous operation. |
| Governed marketplace for spatial perception skills; cascade propagation modeling | Trading perception capabilities between agents under governance. Modeling how physical-world disruptions cascade through the mesh. |
64/055,649 — Bulk-Equipotential Electrochemical Energy Storage Cell with Hydrogen-Activated Metal Nanoflakes in a Dual-Domain Proton-Conducting Carbon Gel
Key disclosed elements
| Disclosed Element | Convergence |
|---|---|
| Bulk-equipotential architecture eliminating internal current collectors; dual-domain proton-conducting carbon gel | Lithium-ion requires metal current collectors, separators, and liquid electrolytes. This eliminates internal metal — the gel itself conducts protons and electrons in separate domains. |
| Hydrogen-activated metal nanoflakes for reversible energy storage within carbon gel matrix | Solid-state hydrogen storage typically requires extreme conditions. This embeds storage at the electrode level within an ambient-condition gel matrix, avoiding lithium supply chain entirely. |
64/070,239 — Agent-Resident Execution Substrate with Governed Inference Tool Registry and Lineage-Derived Personal Corpus Model Training
The integration architecture — where every other filing meets. See the architecture page and white paper.
Claim 1 of 39 (System) — Agent-resident execution substrate
| Element | Convergence |
|---|---|
| Semantic agent as persistent execution substrate surviving device restarts; persistent identity field, cognitive state field, lineage field (append-only), governance policy field | Apple Intelligence resets per-session. ChatGPT memory is prompt replay. No shipping product maintains a durable agent with structurally separated identity, state, lineage, and governance that survives restarts. |
| Tool registry storing managed inference endpoints, each comprising model artifact, interface specification, and governance scope | Ollama, MLX, llama.cpp treat models as files. This manages them as governed assets with specifications and scoped governance. Every OEM shipping multiple on-device models needs this. |
| Agent-to-tool dispatcher routing inference requests based on current cognitive state | LangChain routes by task type. This routes by the agent's cognitive state — context-aware, disposition-aware, readiness-aware tool selection. |
| Tool lifecycle controller performing governed lifecycle operations (installation, retraining, replacement, archival, removal) recorded in lineage | Model lifecycle as a governed, auditable operation. Not "download and run" but "install under policy, retrain under policy, replace under policy, archive under policy." |
| Lifecycle operations do not modify identity or cognitive state; lineage modified only by appending; behavioral continuity independent of loaded model artifacts | The structural separation invariant — the patent's strongest claim. Swap every model, and the agent is still the same agent. No existing system makes this guarantee. This is what blocks design-around. |
Claim 2 of 39 (Method) — Autonomous tool improvement
| Element | Convergence |
|---|---|
| Recording lineage for each inference request: input descriptor, endpoint ID, output descriptor, timestamp, downstream-outcome reference | Observability for inference. Not just logging — structured lineage with downstream-outcome references. The agent tracks whether its tool's outputs actually worked. |
| Detecting tool-improvement trigger: outcome quality threshold, policy-declared schedule, external knowledge ingestion, user request | Four trigger types for retraining. Quality-based, scheduled, knowledge-driven, or user-initiated. The agent decides when its tools need improvement. |
| Deriving supervised training corpus from lineage under corpus policy; fine-tuning targeted endpoint | RLHF/RLAIF retrain from model output (recursive, collapse-prone). This retrains from real-world lineage — actual outcomes, not model opinions. The solution to the model collapse problem. |
| Substituting updated artifact under governed lifecycle; preserving identity, cognitive state, and prior lineage | Hot-swap a retrained model while preserving agent identity. The model improves; the agent remains the same agent. No existing system does this. |
Key dependent claims (37 total)
3 Personal corpus model — endpoint internalizing user's body of work parametrically
4 Personal computing device; user policy as dominant authority
5 Multi-endpoint concurrent routing with aggregation
16 Federation layer preserving local governance sovereignty
18 Cognitive-state dispatch with domain-threshold filtering
20 Multi-deployment: personal, workstation, enterprise, industrial, embedded, robotic, vehicular
26 Hardware-anchored identity binding via device attestation
34 Tool marketplace with publisher attribution and compensation routing
38 Agent lifecycle: snapshot, restore, migration, fork, succession, decommissioning
Portfolio Composition
Designing around one filing pushes you into another
| If You Build... | You Need... | Filings |
|---|---|---|
| A persistent on-device agent | Identity surviving model swaps; unjailbreakable governance; lineage for auditability | 19/538,221 · 19/388,580 · 19/561,229 · 64/070,239 |
| Per-user model specialization | Lineage-derived retraining; identity preservation across fine-tuning; training data admissibility | 64/070,239 · 19/647,395 · PCT/US26/28630 |
| A multi-agent network | Memory-native protocols; decentralized governance; quorum-validated mutations; trust-slope identity | 19/366,760 · 19/326,036 · 19/388,580 · 19/561,229 |
| An AI companion that remembers | Persistent cognitive state with affect and integrity; identity continuity; emotional modeling | 19/647,395 · 19/388,580 · 64/070,239 |
| Autonomous vehicles or robots | Spatial governance; fleet skill emergence; hardware-anchored identity; governed actuation | 64/049,409 · 19/647,395 · 19/388,580 · 64/070,239 |
| Defensible training practices | Per-artifact provenance; rights-grade admissibility; consultation logging | PCT/US26/28630 · 19/647,395 |
| AI safety beyond guardrails | Governance gates; composite admissibility; corrective loops; confidence-governed execution | 19/561,229 · 19/647,395 · 19/230,933 |
AQ
Pre-grant option agreements available.
Claim charts for informational purposes. Claim scope subject to examination. No license granted. Legal.
