Automated Message Resurfacing Based On Interrogative Intent Detection In Electronic Communications, Piyush Goyal

Defensive Publications Series

A common issue in electronic communication is the failure to respond to messages once the unread status is removed. This often occurs when a message containing a direct inquiry is opened but not immediately addressed, causing the communication to be buried by subsequent correspondence. To address this, the disclosed technology employs on-device natural language processing to detect interrogative syntax and intent at the time a message is accessed. If a request or question is identified, a contextual interface element is displayed, allowing for a single-action commitment to follow up. At a scheduled time, the message is automatically resurfaced to the …

The Whisper Web: A Mail Intelligence Framework For Semantic Coordination In Communication Systems, Johny Manuel-Devadoss, Joanna Johnson, Jayden Jack Johnson, Deepthi Prasad

Defensive Publications Series

This paper introduces The Whisper Web, a communication intelligence framework designed for post-static digital correspondence ecosystems operating under conditions of planetary-scale coordination complexity. Unlike conventional email systems that treat messages as isolated immutable payloads exchanged between endpoints, the Whisper Web reconceptualizes communication as a continuously evolving semantic biosphere composed of dynamic probabilistic entities known as Whisper Units. Each Whisper Unit functions as a compressed socio-semantic organism capable of propagating influence through adaptive trust topologies, negotiation-energy fields, obligation gradients, institutional resonance maps, and multi-agent predictive coordination networks.

The proposed framework abandons the traditional paradigms of inboxes, message queues, folders, labels, and …

Gesture Control Of Operations Via Wearable, Russell William Myers, Yeabkal Assegid Wubshit

Defensive Publications Series

This document describes a wearable computing device (e.g., smart watch, smart ring, fitness tracker, hearable device, or other body-mounted devices or sensors) that facilitates gesture control of operations, such as media sessions and applications. For instance, the wearable computing device may allow a user of the wearable computing device to execute gesturebased commands to control a media application or launch specific applications without changing a visual display and/or navigating a menu of the wearable device. The wearable computing device may control media applications across devices (e.g., a video playing on a phone, music playing on headphones, etc.) and monitor gestures, …

Golden Hour Open Conceptual Architecture For Citizen-Initiated Structured Urgency Signaling, Jaime Gerardo Figoni Him

Defensive Publications Series

Abstract

Golden Hour proposes a minimalist, open, and low-bandwidth emergency signaling architecture that enables citizens to transmit structured urgency information during the critical first hour of a disaster. The system combines a single-tap severity code (Red / Yellow / Green), a brief guided question sequence, and resilient transmission paths to create a live, color-coded situational heat map for emergency responders.

By treating emergency response as a physics problem of controlling informational entropy, Golden Hour enables a near-instantaneous reduction in chaos while fostering voluntary human solidarity. The system is designed as a complement to, not a replacement for, existing emergency services.

Llm-Driven Dynamical Adaptation Of Window Layout Based On Available Displays, Aurélien Pomini, Nicolas Zimmermann

Defensive Publications Series

Techniques are described for automatic inference of window layouts based on available displays, e.g., based on detecting connection or disconnection of an auxiliary display. Upon detection of a change in display configuration, contextual information regarding active applications, window metadata, and display parameters is obtained, with user permission. Application categories are determined based on semantic analysis. An on-device large language model is utilized to determine an optimal arrangement of windows across the available displays based on the aggregated context and historical usage patterns. Windows are automatically moved and/or resized according to the inferred layout.

Semantic Photo Search With Compositional Spatial Reasoning, Tamar Yakar, Ilia Labzovsky

Defensive Publications Series

Photo search technologies can identify the content of an image but cannot answer queries relating to the spatial relationships between entities in an image. This disclosure describes computational architectures and techniques for spatial compositional photo search and retrieval. Subjective, spatially rich natural language queries are translated into strict geometric and relational constraints, which are evaluated efficiently against a large corpus of pre-computed spatial indices. Upon ingestion, an image is processed to generate a highly compressed, pre-computed spatial scene graph G = (V, E, C), where V is a set of multimodal entities and bounding boxes; E …

Privacy-Preserving Generative Inpainting, Tamar Yakar, Ilia Labzovsky

Defensive Publications Series

People frequently capture images in public or semi-public settings (e.g., schools, parks, parties) that include incidental subjects and may share such images with others. The person that captures such images may not have consent from such incidental subjects to publish such images. This disclosure describes generative augmentation techniques that preserve privacy for digital media. Incidental subjects (bystanders) in photographs are automatically protected prior to social sharing. Unlike conventional redaction methods (blurring/pixelation) which degrade visual fidelity, or manual object removal which alters the scene's composition, the techniques semantically substitute unauthorized subjects with context-appropriate, non-identifiable synthetic elements. The synthetic elements can be …

Direct-To-Local-Psap Emergency Messaging Over Satellite, Aishwarya Mallampati, Sooraj Sasindran

Defensive Publications Series

When text emergency service (T911) is used over satellite, the carrier configuration may be such that the emergency text is routed to a concierge service that may be outside the country that the user equipment is in. This can result in inefficient emergency response, as there may be a delay in communications to the local public safety answering point (PSAP). This disclosure describes techniques that enable a mobile device to communicate directly with a local PSAP via satellite when terrestrial networks are unavailable. A specialized emergency packet data network (PDN) connection is established that utilizes local breakout (LBO) architecture within …

Memory-Bounded, Provenance-Preserving Canonical Feature Generation For Normalized Entity Keys In Distributed Transaction Aggregation, Arnab Dutta, Rafael Ostrea, Pawan D. Parchani

Defensive Publications Series

The present disclosure relates to a method and a system for generating canonical features for normalized entity keys in distributed transaction aggregation. The present disclosure suggests receiving transaction data comprising a plurality of transaction records, each transaction record associated with an entity identifier derived from one or more entity attributes. After receiving the data, the present disclosure suggests aggregating the transaction records using a normalized merchant entity key, which consolidates multiple entity identifiers corresponding to a same merchant entity. Thereafter, the present disclosure suggests generating an intermediate data structure for each normalized merchant entity key and for each selected attribute …

Artificial Intelligence Social Companion, D Shin

Defensive Publications Series

This disclosure describes an AI-generated social role-play companion that provides emotionally intelligent virtual friends to help users develop social skills through interactive scenarios. Per the techniques, a general-purpose large language model (LLM) is augmented with persona-specific low-rank adapters (LoRA). The LoRA manage multiple AI friend characters, each with distinct personality, communication style, and emotional model. An emotional intelligence module detects the user’s emotional context, using, e.g., voice prosody. A social scenario generator creates age-appropriate social situations focused on, e.g., sharing, conflict resolution, and empathy building. A progress tracker monitors social skill development over time. Aside from character voice responses through …

Personal Significance Scores For Images, Ashwin Goyal

Defensive Publications Series

The true significance of a particular photograph (or video) to a user is difficult to discern in an automatic manner directly from photo pixels (and/or metadata) because there is context beyond the photograph that cannot be determined from pixels alone. This disclosure describes techniques, implemented with user permission, that provide a reliable indicator for the significance of an image using the statistics, telemetry, and metadata relating to action categories such as the capture, curation, consumption, connection, and culling of the image. The techniques generate a personal significance score (PSS) normalized to the range (0.0, 1.0) leverage user behavior with reference …

Acknowledged Queuing Of Sim Proactive Commands By A Mobile Equipment, Pankaj Kumar, Aakash Phoughat

Defensive Publications Series

In mobile telecommunications, a generic busy response from a mobile equipment (ME), such as a smartphone or wearable device, to a proactive subscriber identity module (SIM) command can create ambiguity, potentially causing the SIM to initiate a retry logic. This may lead to redundant command traffic, increased power consumption, and potential race conditions. A signaling mechanism is described that includes a protocol enhancement where an ME can advertise its ability to internally store commands, for example, via a flag in a terminal profile command. When busy, the ME can queue the command and use a specific result code in a …

Dream-State Consolidation Transformers For Long-Horizon Artificial Intelligence, Johny Manuel-Devadoss, Joanna Johnson, Jayden Jack Johnson

Defensive Publications Series

Contemporary transformer architectures achieve strong performance through large-scale parameterization, retrieval augmentation, and extended context windows. However, persistent token-level retention scales poorly for lifelong or long-horizon cognition due to increasing computational cost, retrieval interference, context fragmentation, and memory saturation. This paper proposes a speculative but computationally grounded framework termed Dream-State Consolidation Transformers (DSCT), in which artificial systems alternate between online inference phases and offline consolidation phases. During online operation, systems process high-resolution episodic information from external environments. During offline consolidation, episodic traces are replayed under stochastic perturbation and compressed into semantically stable latent representations.

Unlike conventional persistent-memory systems, DSCT prioritizes abstraction …

Defensive Publication: Ble Peripheral Mitigations For Adaptivity Requirements, Braxton Davis

Defensive Publications Series

A method and system are provided for ensuring regulatory compliance for Bluetooth Low Energy (BLE) peripheral devices operating in a wireless communication environment. The approach addresses scenarios where a central device may not cooperate with peripheral-driven interference reporting or adaptive frequency hopping spread spectrum (AFHSS) requirements. The method involves identifying interfered channels and, for transmissions on such channels, limiting transmissions to short control signals (SCS-Only Transmissions) or adjusting transmit power to a reduced level, such as 10 dBm. For compliant channels, higher transmit power may be maintained. This allows for independent peripheral compliance, improved range, and enhanced link quality without …

Low-Overhead In-Kernel Telemetry For Ml Accelerators Using Shared Memory Fifo, N/A, N/A

Defensive Publications Series

Existing profiling tools for machine learning accelerators often lack the granular visibility to diagnose performance bottlenecks within collective communication kernels, typically identifying only that a kernel is slow without pinpointing the specific cause. A method for in-kernel telemetry uses a shared memory first-in-first-out (FIFO) queue established between a host central processing unit (CPU) and an accelerator device. The accelerator kernel is instrumented with lightweight hooks that write fine-grained telemetry events, such as data arrival timestamps from peer accelerators, directly to the FIFO without blocking execution. A dedicated process on the host asynchronously polls the FIFO to consume the telemetry data. …

Multimodal Intent Prediction Via Retroactive Context Fusion And User Memory Priors, Jinpeng Wang

Defensive Publications Series

Today’s smart devices often feel slow and clunky because they wait for long pauses in speech and ignore what happened just before you started talking. This paper introduces a better way for devices to predict what you want by looking at a mix of clues. The system saves a few seconds of audio before you speak and combines it with where you are looking in real-time. By comparing this info with your past habits, the device can guess your goal before you even finish your sentence. This allows the system to start your request sooner, fix mistakes proactively, and stop …

Real-Time Bio-Adaptive Audio Modification Using Predictive Analysis And Generative Re-Synthesis, Nikhil Kumar

Defensive Publications Series

A described real-time, bio-adaptive audio modification system can operate on a consumer device, such as a smartphone, tablet, or wearable device. The system may use a multi-stage artificial intelligence pipeline to predict potentially startling events from concurrent video and audio streams. It can then use generative source decomposition to isolate a specific sound and apply neural re-synthesis to transform its acoustic characteristics into a form with a potentially reduced physiological impact. System intervention can be dynamically modulated by real-time biometric data from a user, such as heart rate variability, which can provide a proactive, personalized method to reduce physiological responses …

Edge-Computed Dynamic Licensing Using Generative Perception And Agentic Resource Settlement, Nikhil Kumar

Defensive Publications Series

Some digital rights management systems can employ static, user-based licenses that may not fully account for dynamic viewing environments, such as the number of viewers present. An edge computed system can operate locally on a client device, for example, a smartphone, a smart television, or a wearable device. The system may utilize a generative perception engine to analyze multi-modal sensor data, which can generate an anonymized textual description of the viewing scene and perform liveness detection. A dynamic smart contract engine can then synthesize a machine-readable session license based on this real-time contextual information. An autonomous agentic settlement system may …

Cognitive Senate Architecture: A Constitutional Multi-Agent Operating System For Planetary-Scale Ai Coordination, Johny Manuel-Devadoss, Joanna Johnson

Defensive Publications Series

Artificial intelligence systems will no longer exist as isolated language models responding to prompts through stateless inference pipelines. Instead, they will evolve into continuously operating cognitive ecosystems composed of billions of autonomous reasoning entities interacting across planetary infrastructure. These entities will participate in persistent negotiation, distributed planning, memory synchronization, constitutional governance enforcement, economic resource competition, and long-horizon simulation of human civilization itself. The principal technical challenge of this era will not be increasing isolated benchmark intelligence, but maintaining coherent coordination among heterogeneous cognitive actors operating under conflicting objectives, uncertain information, evolving societal norms, finite energy constraints, and dynamically shifting geopolitical …

Gci Vortex-Shield First Responder System (V4.0)

Defensive Publications Series

This disclosure presents a conceptual passive-survivability platform for firefighters and extreme-environment first responders based on vortex-driven thermo-fluidic mechanics, geometry-assisted airflow control, and variable-density structural lattices. The system replaces many conventional powered subsystems with passive mechanical architectures intended to improve reliability, thermal resilience, and operational endurance in high-heat, smoke-dense, and high-noise environments. The platform integrates several geometry-driven subsystems into a unified protective framework. A movement-powered microfluidic cooling garment employs helical J-groove channels and asymmetric flow geometry to circulate cooling fluid through gait-induced compression without electrically driven pumps. The helmet and visor assembly incorporate a multi-layer anisotropic thermal barrier designed to reduce …

Shared Access Token System And Method, Sherwin Somur, Mahmoud Ibrahem Nazmy

Defensive Publications Series

Disclosed herein is a shared access token system and method to enable primary account holders to request and share a shared access token with a recipient. The disclosed system and method provide a secure means to share primary account access, via the shared access token, without a physical card and without directly sharing primary account credentials. Once a request is initiated, the recipient provides authentication, the service provider provisions the shared access token associated with the primary account, the service provider sends the shared access token to a recipient communication device, the service provider activates the shared access token on …

Cryptographic Identity Document System Using Totp- Derived Symmetric Keys For Offline Issuer-Mediated Access Control, Michael Benaudis

Defensive Publications Series

This disclosure describes a cryptographic identity document architecture enabling offline content decryption and issuer-mediated access control through Time-based One-Time Password (TOTP, RFC 6238) derived symmetric keys, without reliance on any server, certificate authority, or pre- shared key material at decryption time. The system combines: (a) dual-domain symmetric key derivation, where two PBKDF2-SHA-256 derived keys (600,000 iterations, NIST SP 800-132 2023) serve respectively for issuer-side storage encryption and recipient-side content decryption; (b) AES-256-GCM authenticated encryption of identity payloads; (c) embedding of an obfuscated TOTP secret in distributed documents using a position-dependent multi-factor XOR transformation that defeats static extraction; (d) IIFE-based JavaScript …

Ghost Developer Protocol (Gdp): A Persistent Intent Architecture For Self-Evolving Software Systems, Johny Manuel-Devadoss, Deepthi Prasad

Defensive Publications Series

Modern AI coding systems are predominantly reactive, prompt-conditioned, session-bound, and human-supervised. They generate outputs in response to explicit requests but do not persistently evolve software systems over time without user initiation. This paper introduces the Ghost Developer Protocol (GDP), a conceptual and architectural framework in which software repositories contain persistent autonomous engineering entities that continuously observe, simulate, and modify systems without external prompts. These entities, referred to as Ghost Developers, form a continuous intent layer embedded within the repository, enabling long-horizon architectural evolution, predictive refactoring, and autonomous system governance. The repository transitions from a passive artifact into an active computational …

Remote Dispatch And Initiation Of Client-Side Agentic Tasks Via Cross-Device Control Protocol, Yijo Liu, Shadia Walsh, Lily Duraiswami, Yeun Su Shin, Tiffany Deonne Morris, Carl Magnus Borg, Golden Gopal Krishna, Clifford Tse Yan Chan

Defensive Publications Series

Architectures for agentic operations often involve a coupling of the input interface and the execution environment on a single piece of hardware. Such configurations can result in resource strain on mobile hardware, affecting battery longevity and data usage, and may encounter compatibility issues with desktop-centric web workflows. A dispatch protocol is disclosed that decouples the input interface from the execution environment. A dispatch payload is generated on a source device (e.g., a smartphone) or relayed via a third-party messaging channel (e.g., enterprise communication platforms or instant messaging applications). This payload includes a task prompt, browser configurations, and user context. The …

Proactive Presentation Of Automated Assistant Prompts Based On Resource Locator Analysis, Yijo Liu, Carl Magnus Borg, Tiffany Deonne Morris, Yeun Su Shin, Shadia Walsh, Clifford Tse Yan Chan, Lily Duraiswami, Golden Gopal Krishna

Defensive Publications Series

A discovery gap exists where users often fail to recognize that an automated assistant is capable of assisting with tasks during a browsing session. This frequently leads to manual navigation through complex web workflows. To address this, a method is disclosed for the proactive presentation of assistant prompts based on resource locator analysis. Upon navigating to a resource locator or receiving a resource locator via a notification, the target page is analyzed to categorize functional capabilities. If a workflow compatible with an automated assistant is identified, such as a multi-step booking or purchasing process, a context-specific prompt is presented. This …

Automated Booking Using Ai-Driven Browser Automation System And Calendar Availability, Benjamin Albert Azose, Jonathan Jerome Azose

Defensive Publications Series

An AI-driven browser automation system is disclosed to autonomously manage reservation and booking tasks by synchronizing real-time web availability with a user’s calendar data. The system navigates booking interfaces, via automated browser interactions or application programming interfaces (APIs), to extract available slots and performs duration-aware compatibility analysis against the user's existing commitments. Supporting both "availability-first" and "calendar-first" processing models, the system identifies non-conflicting intervals to execute bookings in fully automated, semi-automated, or recommendation-based modes. By automating form completion and monitoring for new openings, the framework reduces manual interaction overhead, eliminates scheduling errors, and improves the success rate of securing high-demand …

Automated Generation And Dynamic Insertion Of Image Alternative Text Via Generative Models, Benjamin Albert Azose

Defensive Publications Series

Missing or low-quality alternative text, or alt text, frequently impedes web and document accessibility. Alt text includes descriptive information attached to an image that screen readers use to describe visual content to visually impaired users. A deficiency in alt text creates a barrier for users who depend on screen readers to interpret visual content on webpages. The disclosure describes a method whereby a web browser or document viewer utilizes a generative model to analyze the content of an image to be displayed. If the image is not associated with alternative text that describes the image adequately, the web browser or …

Linguistic Transformation And Multimodal Content Simplification Logic, Robert Gerhard Rudolf Brauer, Isabelle Evelyn Joswig

Defensive Publications Series

A system for linguistic transformation performs modification of an unstructured data payload. The system comprises a content ingestion module, a linguistic transformation module, a multimodal interpreter, and a content output module. The content ingestion module identifies textual elements and non-textual elements within the unstructured data payload. The linguistic transformation module applies standardized parameters to the textual elements, modifying syntax and vocabulary. The multimodal interpreter generates descriptive text strings from the non-textual elements. The linguistic transformation module applies the standardized parameters to the descriptive text strings. The content output module aggregates the modified textual elements and the modified descriptive text strings …

Holographic Data Sharding For Cross-Jurisdictional Storage And Ephemeral Edge Reconstruction, Mohsin Alam

Defensive Publications Series

A waveform security model may use a holographic sharding process to transform a data object into a set of mathematically incomplete and individually unintelligible interference patterns. An orchestration engine can distribute these shards across multiple storage providers and/or jurisdictions, configured to prevent a single entity from possessing sufficient information for reconstruction. The complete data object can be reconstituted, for example, ephemerally within a secure processing environment on an authorized client device (e.g., a smartphone, a laptop computer) at a time of use. This process can render data non-local, disaggregated, and mathematically incomplete at rest, potentially improving data sovereignty and secrecy …

Asynchronous Remote Interaction And Conversational Interrupt Handling For Long-Running Agentic Tasks, Yijo Liu, Yeun Su Shin, Clifford Tse Yan Chan, Shadia Walsh, Lily Duraiswami, Carl Magnus Borg, Tiffany Deonne Morris, Golden Gopal Krishna

Defensive Publications Series

Tasks are managed through a decoupled interaction and execution framework. An execution environment on a primary device performs a task and detects an interrupt state when a decision node is reached that lacks parameters or confirmation. A natural language query object is serialized and routed via a cloud messaging service or chat to an authenticated secondary device. Remote input is captured at the secondary device and transmitted back to the primary device. The input is parsed and injected into the execution runtime, allowing the task to resume from the point of interruption. This process enables the extension of decision-making logic …