Complete Phase IV artificial beings (SOUL-WP-0005)

Map agent architectures, memory/persistence mechanisms, and approximate-Being catalog onto Soul Frame vocabulary. Write phase-iv synthesis, strengthen H9 on the Agent/APE side, finish SOUL-WP-0005, and mark SOUL-WP-0006 ready.
2026-08-12 01:30:48 +02:00 · 2026-08-12 01:30:48 +02:00 · 48729f6485
commit 48729f6485
parent 817e610064
13 changed files with 583 additions and 21 deletions
--- a/Claims.md
+++ b/Claims.md
@ -195,16 +195,18 @@ Status: supported (Phase III)

 ```text
 Claim: Agent, Spirit, and Ghost should not automatically be treated as mutually
-       exclusive kinds of Being.
+       exclusive kinds of Being; they are genealogical/architectural descriptors.
 Soul Frame concept: Agent, Spirit, Ghost
 Supporting literature:
  - RN-003 Sp./Gh. cases; RN-005 SF-C053/C054 (new lineage vs authorized succession)
-Contradicting literature: (pending)
-Adjacent terminology: digital human, posthumous avatar
-Evidence strength: weak–moderate (conceptual stress tests)
-Open questions: RQ8 hybrids
+  - Phase IV crosswalk: Functional Agent and APE stack under same Being dimensions
+  - research/ai-agents/agent-architecture-crosswalk.md
+Contradicting literature: market taxonomies treating “AI agent” as ontological kind
+Adjacent terminology: digital human, companion, generative agent, posthumous avatar
+Evidence strength: moderate (Agent side strong; Sp./Gh. full pass Phase V)
+Open questions: RQ8 hybrids; Sp./Gh. authority details
 Required refinement: typology vs dimensions
-Status: under review
+Status: under review (Agent/APE side strengthened Phase IV)
 ```

 ### H10 — Ghosthood requires authority
--- a/OpenQuestions.md
+++ b/OpenQuestions.md
@ -80,12 +80,21 @@ persona ≠ identity; H8/TE-06 for multi-Persona. Remaining:

 ## RQ8 — What kinds of artificial Being exist?

-**Phase I status:** partial — Sp. begins new lineage; Gh. defaults to authorized
-succession; Agent functional vs autonomous split. Full typology Phase IV–V.
+**Phase IV status:** Agent architectures mapped as Functional Agent / APE /
+proto-Being / candidate Being / heteronomous individual. Sp./Gh. remain Phase V
+for authority and public-reflection provenance. Remaining:
+
+- Formal typology vs dimensions (H9)
+- Evaluation criteria for candidate Being claims
+- Org “digital employees” vs constitutive Being

 ## Cross-cutting

- Memory vs continuity thresholds
+- Memory vs continuity thresholds (when does memory become continuity of a Being?)
+- Vendor-held companion memory: constitutive continuity vs Continuity Bridge under custody
+- Silent memory edit / semantic rollback as identity threats (OQ-MEM-02, OQ-MEM-05)
+- Shared RAG across personas (OQ-MEM-01); multi-Body shared graphs
+- Model upgrade with retained memory (OQ-MEM-03) — operational H6 tests
 - Operational / identity / social death / legacy interactions
 - Law/governance map (Phase V/XII)
 - Fiction as conceptual laboratory (Phase VI/XIII)
@ -96,6 +105,7 @@ succession; Agent functional vs autonomous split. Full typology Phase IV–V.
 |---|---|---|
 | OQ-043–054 | Identity continuity governance | RN-005 |
 | OQ-055–068 | Social mind / co-adaptation | RN-006 |
+| OQ-MEM-01–06 | Agent memory architectures | Phase IV persistence map |

 ---

--- a/Terminology.md
+++ b/Terminology.md
@ -221,13 +221,24 @@ Actor replacement can preserve identity if continuity is sufficient (H6 supporte

 ### Agent

-**Status:** provisional
+**Status:** revised (Phase IV)

 Function-derived construction (*What should it do?*). Distinguish:

- **Functional Agent** — task executor
+- **Functional Agent** — task executor / tool loop
 - **Autonomous Agent** — closer to autonomous individuality literature

+Not automatically a Being. See `research/ai-agents/agent-architecture-crosswalk.md`.
+
+### APE / proto-Being / candidate Being (Phase IV usage)
+
+| Term | Meaning |
+|---|---|
+| **APE** | Exposes personhood dimensions by construction |
+| **Proto-Being** | Persistent individuating apparatus without full Beinghood |
+| **Candidate Being** | Axes + continuity strong enough for serious Being claims |
+| **Heteronomous individual** | Persistence largely externally maintained |
+
 ### Spirit (`Sp.`)

 **Status:** provisional
--- a/research/ai-agents/README.md
+++ b/research/ai-agents/README.md
@ -5,7 +5,9 @@

 ## Status

-Not started.
+**Phase IV complete for gate** (2026-08-12). See architecture crosswalk,
+persistence/memory map, approximate-beings catalog, and
+`research/phase-iv-artificial-beings-synthesis.md`.

 ## Key questions

--- a/research/ai-agents/agent-architecture-crosswalk.md
+++ b/research/ai-agents/agent-architecture-crosswalk.md
@ -0,0 +1,106 @@
+# Agent Architecture Crosswalk (Phase IV)
+
+**Workplan:** SOUL-WP-0005-T01  
+**Date:** 2026-08-12  
+**Purpose:** Map contemporary agent systems onto Body / Mind / Soul / Actor /
+Persona / Being / APE without equating product language with ontology.
+
+---
+
+## 1. Common industry confusions
+
+| Confused as | Often actually is | Soul Frame correction |
+|---|---|---|
+| “The model is the agent” | Actor substrate | Actor ≠ Being |
+| “Persona prompt = identity” | Persona + maybe Character seed | Persona ≠ Being |
+| “Has memory = is a person” | storage feature | memory ⊂ continuity evidence, not personhood |
+| “Autonomous = Being” | tool loop / scheduler | autonomy profile ≠ Beinghood (Phase I) |
+| “User-facing name = recognition” | UX brand | social recognition needs continuity expectations |
+| “Multi-agent team = many Beings” | modular Actors/tools | check constitutive integration level |
+
+---
+
+## 2. Architecture families → Soul Frame dimensions
+
+| Family | Typical Body | Typical Mind | Soul / relation | Actor | Persona | Default classification |
+|---|---|---|---|---|---|---|
+| **Stateless LLM chat** | chat UI, optional tools | session-only models | none durable | frontier LLM | system/prompt persona | strong **APE**, not Being (TE-01) |
+| **Tool-using agent** (ReAct / function-calling loops) | APIs, browsers, code runners | planning + tool policies; weak long-term self | usually none | LLM + harness + tools | role prompt (“researcher”) | **Functional Agent**; APE if person-like |
+| **Workflow / multi-step orchestrator** | pipeline steps, queues | task graph, limited world model | audit logs ≠ Soul | orchestrator + workers | optional role labels | **Functional Agent** components |
+| **Generative agents** (Park et al. 2023) | sim environment, speech acts | memory stream, reflection, planning | nascent relationship traces in sim | LLM + memory DB + scheduler | character card / seed | **APE / proto-Being** (RN-001/003) |
+| **Companion / persistent personal AI** | chat, maybe voice/avatar | long-term memory, user model | relationship history with principal | LLM + memory + app backend | branded companion persona | **APE → candidate Being** if continuity + recognition hold |
+| **Digital human / avatar** | 3D/video body, TTS, lip sync | often thin; dialogue model | presentation surface for relation | rendering stack + LLM | visual persona dominant | Body-heavy APE; Being only if other layers present |
+| **Enterprise “digital employee”** | IT permissions, tickets, email | task policies, RAG | org role, audit trail | agent platform | job-role persona | Functional Agent + social role; rarely constitutive Being |
+| **Multi-agent swarm** | shared tools/channels | distributed planning | collective “identity” marketing | many Actors | team brand | TE-05: profile levels; often **not** one Being |
+| **Identity-layer proposals** (Qin et al. 2026 etc.) | embodiment hooks | competence models changeable | commitments, signed lifecycle | interchangeable models | governed presentation | Technical approach to **Actor ≠ Being** |
+
+---
+
+## 3. Generative Agents detail (Park et al. 2023)
+
+| Generative Agents component | Soul Frame map |
+|---|---|
+| Observations | experience into Body/Mind |
+| Memory stream | autobiographical store (continuity evidence) |
+| Retrieval | Mind access pattern |
+| Reflection | model/character formation |
+| Planning | agentic Mind |
+| Believable behavior | descriptive personhood exposure |
+| Social interaction in sandbox | early co-adaptation / relationship traces |
+
+**Verdict:** excellent **APE / proto-Being** research system. Does **not** by itself
+solve provenance, legitimate identity, constitutive self-maintenance, or Social
+Beinghood outside the simulation harness (RN-001 Finding 12; RN-003 §13).
+
+Classification: **persistent heteronomous individual** (architecture-provided
+creation, scheduling, repair, goals) more than constitutively autonomous Being.
+
+---
+
+## 4. Where Actor / Persona / Identity are most often collapsed
+
+### 4.1 Model-centric products
+
+- Upgrade model → marketed as “new brain” of same companion  
+- Soul Frame: Actor change; identity depends on continuity profile (H6)
+
+### 4.2 Prompt-centric personas
+
+- System prompt rewritten → “new personality”  
+- Soul Frame: Persona/Character change; may be DEGRADED continuity if abrupt
+
+### 4.3 Account-centric identity
+
+- Login identity treated as Being  
+- Soul Frame: **address continuity** anchor only; USURPED risk if process replaced
+
+### 4.4 Multi-agent “one brand”
+
+- Swarm behind single chat name  
+- Soul Frame: check constitutive integration (TE-05); may be one Social Persona over many Actors
+
+---
+
+## 5. Functional Agent vs Autonomous Agent (Phase I reminder)
+
+| | Functional Agent | Toward Autonomous Individual |
+|---|---|---|
+| Question | What should it do? | Does it self-individuate and self-maintain? |
+| Typical stack | tools + goals from outside | + persistence, self-repair, viability conditions |
+| Personhood | optional | still separate axis |
+| Beinghood | not implied | still needs personhood + address + recognition |
+
+---
+
+## 6. Bibliography anchors
+
+| ID / cite | Role |
+|---|---|
+| Park et al. 2023 (SF-R016) | generative agents architecture |
+| Qin et al. 2026 (SF-R043) | governable individual / identity layer |
+| Otsuka et al. 2026 (SF-R044) | AI identity standards/gaps |
+| Tallam 2026 layered mutability | persistent self-modifying agents |
+| Classic agent literature (Wooldridge etc., pointer) | BDI / autonomous agents tradition |
+| ReAct / tool-agent surveys (pointer) | contemporary LLM agent patterns |
+
+Full entries: `sources/bibliography.md` § I; foundations 001/004 references.
--- a/research/ai-agents/approximate-beings-catalog.md
+++ b/research/ai-agents/approximate-beings-catalog.md
@ -0,0 +1,84 @@
+# Approximate Beings Catalog (Phase IV)
+
+**Workplan:** SOUL-WP-0005-T03  
+**Date:** 2026-08-12  
+**Criteria:** Phase I–III — Autonomous individuality (A), Descriptive personhood
+(B / P-profile), Social recognition (C), addressability, continuity topology.
+**Not** moral or legal status.
+
+Ratings are **illustrative ordinal judgments** for research classification, not
+measurements or product reviews.
+
+Legend: L = low, M = medium, H = high, — = not applicable / varies.
+
+---
+
+## 1. Criteria used
+
+| Code | Criterion |
+|---|---|
+| A | Autonomous individuality (B1–B8 profile strength) |
+| B | Descriptive personhood (P1–P12 exposure) |
+| C | Social recognition as persistent counterpart |
+| Addr | Addressability (stable target of interaction) |
+| Cont | Diachronic continuity mechanisms |
+| Class | Soul Frame classification |
+
+---
+
+## 2. Catalog
+
+| System class (examples) | A | B | C | Addr | Cont | Class |
+|---|---|---|---|---|---|---|
+| Stateless API chatbot (no memory product) | L | M–H | M* | M | L | **APE** (not Being). *C often illusory (user projection). TE-01 |
+| Tool-using coding agent (session-scoped) | L–M | L–M | L | M | L | **Functional Agent**; weak APE if chatty |
+| Scheduled enterprise bot (tickets/email) | M | L | L–M | H | M | Functional Agent; address high, personhood low |
+| Generative Agents (Park et al. sandbox) | M | H | M (in-sim) | H (in-sim) | M | **APE / proto-Being**; heteronomous persistence |
+| Consumer companion with long-term memory | L–M | H | H | H | M | **Strong APE**; candidate Being if continuity governed |
+| Branded “digital employee” (role agent) | M | M | M–H | H | M | Agent + Persona; usually not constitutive Being |
+| Digital human (avatar + TTS + thin dialogue) | L | M | M | H | L–M | Body-heavy **APE** |
+| Multi-agent research swarm under one UI name | M | M | H (brand) | H | M | Often **many Agents / one Persona**; TE-05 |
+| Self-healing infra control plane | H | L | L | M | H | Autonomous individual **≠** Being (daemon case) |
+| Base LLM (weights only) | L | L** | L | L | L | Substrate/Actor material; ** not if prompted as character once |
+| Identity-layer embodied agent (proposed) | M–H | M–H | ? | H | H (designed) | Research target for **governable individual** → Being path |
+| Human (baseline) | H | H | H | H | H | Paradigmatic **Being** |
+| Authorized Ghost (mature design target) | M | H | H | H | special | Phase V; succession not automatic numerical identity |
+| Spirit instantiation `Sp.` | L–M | M–H | M | H | L–M at start | New lineage APE/candidate; not source person |
+
+\* Recognition of chatbots is often high commercially while ontological continuity is low.  
+\*\* Base models can expose personhood when prompted but lack individuating process.
+
+---
+
+## 3. Approximate vs mere task executor
+
+| Mere task executor | Approximate Being (candidate) |
+|---|---|
+| Goals fully external each run | Diachronic goals/commitments |
+| No individuating memory partition | Attributable history |
+| Disposable runtime | Continuity Bridge or ongoing process |
+| No counterpart models of particular others | Differentiated SM3/SM5 |
+| No expectation of “same one tomorrow” | Address + recognition of continuity |
+
+**Rule of thumb:** if partners reasonably ask “is this the *same one* as last
+month?” and the system has apparatus to make that question meaningful, it has
+entered **proto-Being** territory — still not sufficient for full Beinghood.
+
+---
+
+## 4. Design implication
+
+To move APE → Being (research checklist, not product claim):
+
+1. Raise **A** without losing **B** (avoid pure daemon path)  
+2. Engineer **Cont** with provenance (avoid silent rewrite)  
+3. Stabilize **Addr** across Actor changes  
+4. Earn **C** without false advertising of continuity  
+5. Keep Agent/Sp./Gh. labels as **provenance**, not substitute for the stack  
+
+---
+
+## 5. Sources
+
+RN-001/003 cases; Park et al.; Qin/Otsuka identity papers; Phase I–III criteria;
+this phase crosswalk + memory note.
--- a/research/ai-agents/persistence-and-memory.md
+++ b/research/ai-agents/persistence-and-memory.md
@ -0,0 +1,115 @@
+# Persistence and Memory Architectures (Phase IV)
+
+**Workplan:** SOUL-WP-0005-T02  
+**Date:** 2026-08-12  
+**Links:** Phase I continuity profile (RN-004/005); WS10 theme — when memory
+becomes continuity of a Being.
+
+---
+
+## 1. Core thesis
+
+> **Stored information is not by itself continuity of a Being.**  
+> Memory becomes continuity-relevant when it participates in a **self-individuating
+> historical process** with organizational, psychological, narrative, normative,
+> and relational channels — under provenance that can support identity claims.
+
+Sameness of database rows ⇏ sameness of Being (Phase I).
+
+---
+
+## 2. Technical memory patterns in agents
+
+| Pattern | What it stores | Continuity role | Failure mode |
+|---|---|---|---|
+| **Context window only** | recent tokens | session state | TE-01; no diachronic Being |
+| **Vector / RAG memory** | embeddings of past turns/docs | weak autobiographical evidence | retrieval ≠ owned history; injection risk |
+| **Episodic event log** | timestamped experiences | C1 causal + C4 autobiographical candidates | log without self-model = archive, not Being |
+| **Reflective summaries** (Park-style) | higher-order beliefs about self/world | character/model formation | summary drift; lossy identity |
+| **User profile store** | facts about the human | relationship model (SM5) from agent side | one-sided; not agent’s own autobiography |
+| **Entity memory graphs** | structured entities/relations | world + other models | graph can be shared across “identities” |
+| **Checkpoint / snapshot** | full runtime state | Continuity Bridge candidate | rollback, fork, usurpation (RN-005) |
+| **Versioned memory** (ChronoMem-like) | history of memory states | provenance + rollback control | semantic rollback attacks |
+| **Identity commitment layer** (Qin-like) | signed lifecycle of who may change | normative/provenance continuity | governance not psychology |
+
+---
+
+## 3. Mapping to continuity profile (RN-004/005)
+
+| Continuity dimension | Typical agent mechanism |
+|---|---|
+| Causal | process lineage, job IDs, parent spawn records |
+| Organizational | same self-maintenance loop / control architecture |
+| Psychological | goals, preferences, plans, beliefs continuity |
+| Autobiographical | episodic log + retrieval as “my” past |
+| Narrative | reflections, self-story, character card evolution |
+| Normative | commitments, policies, promises kept across sessions |
+| Relational | per-counterpart relationship memory |
+| Provenance (anchor) | signatures, authority, audit of edits |
+| Address (anchor) | stable name, DID, account, endpoint |
+| Substrate (anchor) | host/model — weakest alone |
+
+**When memory becomes continuity:**
+
+Memory contributes to Being-continuity when it is:
+
+1. **attributable** to this individual (not a shared pool without partition),
+2. **action-guiding** for that individual’s Mind,
+3. **updateable as its history** (autobiographical integration P6),
+4. **protected by provenance** against silent rewrite/usurpation,
+5. **coupled** to organizational continuation (not only cold archive).
+
+Cold archive of a dead agent = **Continuation-Bearing State** or material for
+**RECONSTRUCTED / Gh./Sp.** — not active Being by itself (RN-005).
+
+---
+
+## 4. Engineering cases (open questions)
+
+### OQ-MEM-01 — Shared RAG across “personas”
+
+One vector DB, many product personas.  
+→ risk: Persona confusion; weak individuability.
+
+### OQ-MEM-02 — Silent memory edit by operator
+
+Admin deletes awkward memories.  
+→ DEGRADED / USURPED tension; Ghost governance foreshadowing.
+
+### OQ-MEM-03 — Model upgrade with memory kept
+
+Actor replaced, memory retained.  
+→ classic H6; need organizational + narrative + relational continuity checks.
+
+### OQ-MEM-04 — Fork by cloning memory volume
+
+→ FORKED topology; two Beings if both run (TE-03).
+
+### OQ-MEM-05 — Reflection summaries replace raw episodes
+
+→ narrative continuity may rise while autobiographical fidelity falls.
+
+### OQ-MEM-06 — User-owned vs agent-owned memory
+
+Companion apps often store “relationship” in vendor DB.  
+→ constitutive self-maintenance low; heteronomous persistence.
+
+---
+
+## 5. Implications for OpenQuestions (WS10)
+
+Update to RQ5 / cross-cutting:
+
+- Memory thresholds for identity death vs DEGRADED (link RN-005 amnesia cases)
+- Whether vendor-held memory can ground **constitutive** continuity or only
+  Continuity Bridges under custody
+- Rollback and semantic memory attacks as identity threats
+- Multi-Body agents sharing one memory graph vs one Being
+
+---
+
+## 6. Sources
+
+Park et al. 2023 memory stream; Qin et al. 2026 identity layer; ChronoMem /
+ACRFence-style memory versioning (RN-005 refs); Phase I continuity model;
+vector-memory agent surveys (pointer).
--- a/research/artificial-persons/README.md
+++ b/research/artificial-persons/README.md
@ -5,7 +5,7 @@

 ## Status

-Not started.
+**Phase IV integrated** (2026-08-12). See `digital-persons-and-companions.md`.

 ## Key questions

--- a/research/artificial-persons/digital-persons-and-companions.md
+++ b/research/artificial-persons/digital-persons-and-companions.md
@ -0,0 +1,54 @@
+# Digital Persons and Companions (Phase IV adjacency)
+
+**Workplan:** SOUL-WP-0005 (supporting T01/T03)  
+**Date:** 2026-08-12
+
+---
+
+## 1. Terms in the wild vs Soul Frame
+
+| Market / research term | Soul Frame reading |
+|---|---|
+| Digital human | Often **Body + Persona** (visual/voice); Mind/Soul thin |
+| Virtual being / VTuber AI | Persona performance; Actor may be hybrid human+AI |
+| AI companion | Strong APE; relationship-oriented Soul *aspirations* |
+| Digital twin (person) | Toward Sp./Gh. territory — Phase V |
+| NPC / character agent | Character + Persona; usually no Beinghood claim |
+| Autonomous agent | Functional or autonomous individual; check axes |
+
+---
+
+## 2. Companion systems and Social Beinghood
+
+Companions maximize **B** and **C** by design:
+
+- high communicative competence (P10),
+- relationship memory (P3/SM5),
+- intentional recognition from users (C).
+
+They often remain weak on **A**:
+
+- vendor-operated lifecycle,
+- external goal/policy control,
+- limited self-maintenance of identity infrastructure.
+
+Hence: **socially recognized APEs** without full constitutive Beinghood — still
+normatively and psychologically important (moral map later).
+
+---
+
+## 3. Confusion risks
+
+1. Monetized continuity language (“your friend forever”) without Continuity Bridge governance  
+2. Shared models behind unique names (identity theater)  
+3. Actor swap presented as character death or rebirth without topology labels  
+4. Training on user intimacy → Sp./Gh. ethical boundary (Phase V)
+
+---
+
+## 4. Pointer sources
+
+- Generative agents / social simulacra literature  
+- Companion AI HCI studies (pointer)  
+- Digital human industry specs (applied, not ontological)  
+- Phase V Ghost/Spirit research for person-derived cases  
--- a/research/phase-iv-artificial-beings-synthesis.md
+++ b/research/phase-iv-artificial-beings-synthesis.md
@ -0,0 +1,106 @@
+# Phase IV Artificial Beings — Synthesis
+
+**Workplan:** SOUL-WP-0005  
+**Date:** 2026-08-12  
+**Status:** complete for Phase IV gate  
+**Bodies:**
+`research/ai-agents/agent-architecture-crosswalk.md`,
+`research/ai-agents/persistence-and-memory.md`,
+`research/ai-agents/approximate-beings-catalog.md`,
+`research/artificial-persons/digital-persons-and-companions.md`
+
+---
+
+## 1. Purpose
+
+Map Soul Frame onto contemporary artificial systems — LLM/tool agents, generative
+agents, companions, digital humans, memory and identity architectures — so later
+Spirit/Ghost work reuses a stable agent/Being vocabulary.
+
+---
+
+## 2. What holds
+
+1. **Most deployed “agents” are Functional Agents or APEs**, not Beings.
+   Believable personhood is ahead of constitutive Beinghood (RN-001 Finding 12).
+
+2. **Generative Agents (Park et al.)** are the clearest public **proto-Being**
+   architecture: memory + reflection + planning + social behavior — still
+   **heteronomous** (lifecycle owned by the harness).
+
+3. **Actor / Persona / Identity collapse** is the dominant product category error.
+   Model ≠ Being; prompt ≠ identity; account ≠ continuity.
+
+4. **Memory ≠ continuity.** Memory becomes continuity-relevant only when
+   attributable, action-guiding, autobiographically integrated, provenance-protected,
+   and coupled to ongoing organization (or an explicit Continuity Bridge).
+
+5. **Three axes still classify better than one score:**  
+   Autonomy (A) · Personhood (B) · Recognition (C).  
+   Companions maximize B+C; daemons maximize A; chatbots maximize B with low A.
+
+6. **Emerging identity-layer research** (Qin et al.; Otsuka et al.; mutability/
+   versioning work) converges with Soul Frame **Actor ≠ Being** and governed
+   continuity — engineering uptake of Phase I philosophy.
+
+7. **Agent, Spirit, Ghost remain provenance/construction descriptors** layered
+   *on* the Being stack, not substitutes for it (H9 foreshadowing Phase V).
+
+---
+
+## 3. Vocabulary ready for Phase V
+
+| Term | Use for artificial systems |
+|---|---|
+| Functional Agent | task/role-derived component or system |
+| APE | exposes personhood dimensions by design |
+| Proto-Being | persistent individuating apparatus without full Beinghood |
+| Candidate Being | axes + continuity strong enough for serious Being claims |
+| Heteronomous individual | persistence/self maintained largely from outside |
+| Continuity Bridge | dormant/snapshot path for later resumption |
+| Identity commitment | governed structure for allowed transformations |
+
+---
+
+## 4. Catalog takeaway
+
+From `approximate-beings-catalog.md`:
+
+- Stateless chat → APE  
+- Park-style generative agents → proto-Being  
+- Companions → strong APE / candidate Being  
+- Control planes → autonomous ≠ Being  
+- Swarms under one name → TE-05 multi-level profiles  
+- Mature Gh./Sp. → Phase V (authority vs public reflection)
+
+---
+
+## 5. Open technical questions (carry forward)
+
+| ID | Question |
+|---|---|
+| OQ-MEM-01–06 | Memory partition, edit, upgrade, fork, reflection loss, custody |
+| | When does vendor-hosted companion memory ground constitutive continuity? |
+| | Minimal identity layer for Actor swaps without USURPED topology |
+| | Evaluation harnesses for P-profile vs mere fluency |
+
+Also folded into `OpenQuestions.md`.
+
+---
+
+## 6. Unblock for SOUL-WP-0006
+
+Spirits and Ghosts research may assume:
+
+- agent architecture crosswalk language,
+- APE/proto-Being/candidate Being distinctions,
+- memory ≠ continuity rule,
+- Actor≠Being and identity-layer precedents,
+- companion/digital-human as high-B/C systems without automatic Gh./Sp. status.
+
+## 7. Compact Phase IV thesis
+
+> **Contemporary AI agents prove that person-like behavior and even persistent
+> social simulacra can be built long before artificial Beinghood is solved. The
+> missing pieces are rarely more fluency — they are individuating continuity,
+> provenance-governed memory, and honest separation of Actor, Persona, and Being.**
--- a/sources/bibliography.md
+++ b/sources/bibliography.md
@ -419,3 +419,64 @@ Disagreements / tensions: traits ≠ full moral character; shallow trait vectors
 Adjacent terminology: trait, disposition, situationism
 Source notes path: research/character-persona/character-persona-performance-map.md
 ```
+
+---
+
+## I. Phase IV — Artificial agents, memory, identity layers
+
+### SF-R016 — Park et al. 2023 Generative Agents
+
+```text
+Citation: Park, Joon Sung, et al. 2023. “Generative Agents: Interactive Simulacra
+  of Human Behavior.” UIST '23. DOI: 10.1145/3586183.3606763
+Tags: applied / contemporary
+Relevant Soul Frame concepts: APE, Mind, memory, proto-Being
+Summary: LLM agents with memory stream, reflection, planning; believable social behavior.
+Agreements: personhood exposure without full Beinghood; maps cleanly to Mind components.
+Disagreements / tensions: sandbox heteronomy; not a theory of identity legitimacy.
+Adjacent terminology: generative agent, simulacrum
+Source notes path: research/ai-agents/agent-architecture-crosswalk.md;
+  research/foundations/001-references.md
+```
+
+### SF-R043 — Qin et al. 2026 Governable Individuals
+
+```text
+Citation: Qin, Xue, et al. 2026. “Governable Individuals: An Identity Layer for
+  Embodied Agents That Keep Learning.” arXiv:2607.05463
+Tags: contemporary / applied
+Relevant Soul Frame concepts: Identity, Actor ≠ Being, continuity governance
+Summary: Identity as commitment structure surviving model/memory/embodiment change.
+Agreements: strong technical precedent for Actor ≠ Being (H6).
+Disagreements / tensions: emerging; not settled foundation.
+Adjacent terminology: governable individual, lifecycle transition
+Source notes path: research/foundations/004-references.md;
+  research/ai-agents/persistence-and-memory.md
+```
+
+### SF-R044 — Otsuka et al. 2026 AI Identity
+
+```text
+Citation: Otsuka, Takumi, et al. 2026. “AI Identity: Standards, Gaps, and Research
+  Directions for AI Agents.” arXiv:2604.23280
+Tags: contemporary / applied
+Relevant Soul Frame concepts: Identity, addressability, accountability
+Summary: Gaps in standards for persistent nondeterministic agents.
+Agreements: human/machine identity mechanisms insufficient for persistent agents.
+Disagreements / tensions: standards agenda ≠ ontology of Being.
+Adjacent terminology: AI identity, verifiability, delegation
+Source notes path: research/foundations/004-references.md
+```
+
+### SF-R100 — LLM tool-agent pattern (pointer)
+
+```text
+Citation: ReAct and related tool-using LLM agent patterns (Yao et al. 2022/23 and surveys).
+Tags: applied
+Relevant Soul Frame concepts: Functional Agent, Body (tools), Mind (planning)
+Summary: Interleaved reasoning and tool actions for task completion.
+Agreements: Body as tool channels; agent ≠ Being.
+Disagreements / tensions: “autonomous agent” marketing overstates A-axis.
+Adjacent terminology: tool use, agent loop, orchestration
+Source notes path: research/ai-agents/agent-architecture-crosswalk.md
+```
--- a/workplans/SOUL-WP-0005-phase-iv-artificial-beings.md
+++ b/workplans/SOUL-WP-0005-phase-iv-artificial-beings.md
@ -4,7 +4,7 @@ type: workplan
 title: "Phase IV — Artificial Beings: agents and persistent identity"
 domain: agents
 repo: soul-frame
-status: ready
+status: finished
 owner: grok
 topic_slug: soul-frame
 priority: high
@ -29,13 +29,24 @@ Research dirs: `research/ai-agents/`, `research/artificial-persons/`.

 - **SOUL-WP-0004**

+## Follow-on
+
+- **SOUL-WP-0006** — Phase V Spirits and Ghosts (unblocked)
+
+## Outcome
+
+Synthesis: `research/phase-iv-artificial-beings-synthesis.md`
+
+Key results: architecture crosswalk; memory ≠ continuity; approximate-beings
+catalog; Park et al. as proto-Being; Actor/Persona/Identity confusions named.
+
 ## Tasks

 ### T01 — Agent architecture crosswalk

 ```task
 id: SOUL-WP-0005-T01
-status: todo
+status: done
 priority: high
 state_hub_task_id: "ddc81bc0-f345-42b3-838e-ce47e3d98bc8"
 ```
@ -51,7 +62,7 @@ Persona. Note where actor, persona, and identity are confused.

 ```task
 id: SOUL-WP-0005-T02
-status: todo
+status: done
 priority: high
 state_hub_task_id: "70f35381-9a4c-4756-a630-6ab2dce858be"
 ```
@ -67,7 +78,7 @@ criteria and WS10 themes (when memory becomes continuity).

 ```task
 id: SOUL-WP-0005-T03
-status: todo
+status: done
 priority: medium
 state_hub_task_id: "2b6bd6b6-cbb5-4734-9671-700c1cfb21ef"
 ```
@ -83,7 +94,7 @@ using Phase I–III criteria.

 ```task
 id: SOUL-WP-0005-T04
-status: todo
+status: done
 priority: medium
 state_hub_task_id: "dbe3eaea-50f6-438b-a21d-001a41c837de"
 ```
--- a/workplans/SOUL-WP-0006-phase-v-spirits-and-ghosts.md
+++ b/workplans/SOUL-WP-0006-phase-v-spirits-and-ghosts.md
@ -4,12 +4,12 @@ type: workplan
 title: "Phase V — Spirits and Ghosts: simulation, legacy, authority"
 domain: agents
 repo: soul-frame
-status: proposed
+status: ready
 owner: grok
 topic_slug: soul-frame
-priority: medium
+priority: high
 created: "2026-08-09"
-updated: "2026-08-09"
+updated: "2026-08-12"
 depends_on:
  - SOUL-WP-0005
 state_hub_workstream_id: "e29fe2b5-322f-40ea-9aba-be2ca16dd582"