PDF page 1
Object-Bound Agentic Interfaces
Receiver-First Spatial Inventories for Post-Smartphone Computing
DOI: 10.5281/zenodo.19500161
Raynor Eissens · Ambient Era Canon · 2026
⸻
Abstract
This paper introduces Object-Bound Agentic Interfaces (OBAI): a post-smartphone interface
paradigm in which physical objects function as primary interface anchors, activating spatially
bound inventories that contain app shortcuts, dynamic state signals, and agent-generated
outputs.
In contrast to app-centric and chat-centric interaction models, OBAI operates under a receiver-
first logic:
object → receive → generate
AI output is not immediately generated in abstract interfaces, but is instead triggered by
interaction with a real-world object, and may be persistently placed back into that object’s
spatial inventory.
This paper argues that existing AR systems, object recognition tools, and spatial computing
platforms provide partial precedents, but do not implement:
• unified object-bound inventories
• persistent, fading state layers
• integration of apps and agent outputs
• autonomous agentic “landing” of outputs into object-specific contexts
OBAI defines a new interface primitive:
the object as a living, persistent, spatial interface container
⸻
PDF page 2
Canonical Definition
Object-Bound Agentic Interfaces are spatial computing systems in which physical objects act as
persistent interface anchors that activate object-specific inventories containing app shortcuts,
chromatic state signals, and agent-generated outputs, with generation occurring only after
object-based reception.
⸻
Core Principle
Traditional model:
input → generate → display
Object-Bound model:
object → receive → contextualize → generate → land
The object is not the target of output.
The object is the condition for output.
⸻
Problem Statement
Contemporary AI and interface systems exhibit three dominant limitations:
1. Interface detachment
Interaction occurs in abstract containers (apps, chats, dashboards)
disconnected from physical context.
2. Immediate generation bias
AI generates output without environmental grounding, leading to
overload, irrelevance, or instability.
3. Non-persistent contextualization
Outputs are ephemeral and not anchored to meaningful real-world
structures.
Existing systems increase capability, but lack situated coherence.
As identified in prior work, intelligence without environmental support
leads to pressure accumulation and instability .
⸻
PDF page 3
Proposed Architecture
OBAI introduces a three-layer system:
1. Object Layer (Receiver Layer)
Physical objects act as:
• entry points
• contextual anchors
• identity surfaces
Examples:
• PlayStation → gaming context
• refrigerator → consumption/logistics
• plant → care/temporal cycle
• bag → movement/preparation
⸻
2. Spatial Inventory Layer
Each object activates a dedicated inventory interface containing:
• app shortcuts (object-relevant utilities)
• chromatic state markers (living signals, fading over time)
• contextual actions
• agent-generated outputs
This inventory is:
• spatially anchored
• persistent
• dynamically evolving
⸻
PDF page 4
3. Agentic Layer
Agents operate as:
• detectors (events, updates, signals)
• interpreters (context relevance)
• producers (outputs, suggestions, actions)
Crucially:
Agents do not output globally.
They land outputs into object-specific inventories.
⸻
Interaction Model
Minimal loop:
scan → reveal → select → generate → land → fade
Expanded:
1. User observes or scans object
2. Object activates spatial inventory
3. User selects or inspects chroma/app
4. AI generates context-specific output
5. Output is placed back into inventory
6. State decays over time (fade / afterfield)
This extends the ARC-1 logic of field-first interaction and afterfield
decay into a general interface paradigm .
⸻
PDF page 5
Receiver-First Logic
The defining inversion:
Generation is not primary.
Reception is primary.
AI output is:
• delayed until context exists
• grounded in object presence
• spatially returned to that context
This resolves:
• interface overload
• notification drift
• context fragmentation
⸻
Relation to Prior Art
Closest precedents include:
• object-centric AR interaction systems
• spatial UI anchored to surfaces
• multimodal AI with contextual outputs
However, no identified system combines:
• persistent object-bound inventories
• app + agent unification
• autonomous output landing
• receiver-first generation logic
Thus:
Partial prior art only.
The novelty lies in the integration and structural coupling of these elements.
⸻
PDF page 6
System Properties
OBAI systems exhibit:
• situated intelligence
• persistent context memory
• low-symbolic signaling (chroma, fade, presence)
• environmental UI distribution
• non-intrusive output delivery
⸻
Conceptual Shift
From:
• app-centric computing
• feed-based interaction
• notification systems
• chat-based AI
To:
• object-centric computing
• environment-bound interaction
• ambient state signaling
• agentic contextual landing
⸻
Example
Object: PlayStation
Inventory contains:
• game shortcuts
• friend presence indicators
• agent-generated recommendations
• call actions
• live chromatic states
Agent detects new JRPG →
lands result as chroma →
user opens →
PDF page 7
content generated on demand
No global notification required.
⸻
Relation to Reasoning Systems
OBAI operates as an interface layer.
It may be supported by underlying reasoning systems such as structured routing architectures
(e.g. operator-based reasoning stacks) that refine output prior to externalization .
However, OBAI itself defines:
where output appears, not how reasoning is performed.
⸻
Why It Matters
As AI becomes agentic, persistent, and ambient:
• output volume increases
• context fragmentation increases
• user overload increases
OBAI introduces:
environment as interface
object as anchor
presence as filter
This transforms AI from:
a system that produces outputs
into:
a system that places meaning where it belongs
⸻
PDF page 8
Conclusion
Object-Bound Agentic Interfaces define a new interface primitive:
the physical object as a living, spatial, persistent interface
They resolve key limitations in current AI interaction models by:
• grounding output in context
• delaying generation until reception
• distributing interaction across the environment
The result is a system where:
interfaces are not opened
but
revealed in place
⸻
Keywords
object-bound interface, spatial computing, agentic AI, ambient interface, AR interaction,
receiver-first systems, contextual AI, chromatic interface, spatial inventory, post-smartphone UI,
ambient era
⸻
One-Sentence Version
Physical objects become the primary interface, and AI outputs land where they are needed
instead of appearing everywhere.