# AKI™ AI User Behaviour & Delegation Intelligence Atlas (ZIP-1.0)
> Empirical human-AI interaction telemetry auditing 10 enterprise workload domains, the 6-tier Delegation Ladder (D0–D5), 7 semantic prompt intents, cross-model switching matrices, and cognitive reliance radars under differential privacy (ε ≤ 0.5, k ≥ 1,000).

## Macro Actuarial Telemetry:
- Audited Interaction Sessions: NaNM
- Tracked Enterprise Workload Domains: undefined
- Mean Enterprise Delegation Ladder Tier: undefined
- Macro Autonomous Delegation Share: undefined%
- Macro Terminal / CLI Adoption: undefined%
- Macro Unreviewed Diff Acceptance Rate: undefined%
- Fastest Displacing Domain: undefined
- Primary Model Churn Driver: undefined
- Most Fragile Workflow: undefined
- Differential Privacy Guarantee: undefined
- Actuarial Protocol: undefined
- Last Audited (UTC): 2026-09-08T22:00:00Z

---

## 1. Tracked Workload Domains & Displacement Velocity:
| Domain | Category | Median Turn Depth | Primary Intent | Autonomous Share % | Terminal CLI % | Human Intervention % | Velocity Tier |
| :--- | :--- | :---: | :--- | :---: | :---: | :---: | :---: |
| **Software Engineering & Architecture** | Technical Engineering | 14.8 turns | `Delegate & Automate` | **72.8%** | 68.4% | 27.2% | `HYPER_VELOCITY` |
| **Enterprise IT & DevOps SRE** | Infrastructure & Systems | 18.5 turns | `Provision & Remediate` | **81.2%** | 86.7% | 18.8% | `HYPER_VELOCITY` |
| **Customer Support & Case Resolution** | Operations & Support | 3.8 turns | `Deflect & Resolve` | **76.4%** | 18.2% | 23.6% | `HYPER_VELOCITY` |
| **Financial Modeling & Quantitative Audit** | Finance & Accounting | 12.6 turns | `Calculate & Backtest` | **69.5%** | 52.4% | 30.5% | `HYPER_VELOCITY` |
| **Retail Search & Conversational Buying** | Consumer & Commerce | 4.6 turns | `Filter & Purchase` | **62.4%** | 9.8% | 37.6% | `HIGH` |
| **Legal M&A & Statutory Analysis** | Legal & Regulatory | 8.4 turns | `Analyze & Redline` | **54.2%** | 21% | 45.8% | `HIGH` |
| **Academic Tutoring & STEM Research** | Education & Science | 11.2 turns | `Derive & Solve` | **58.1%** | 38.5% | 41.9% | `HIGH` |
| **Content Creation & Media Production** | Creative & Marketing | 7.4 turns | `Generate & Polish` | **64.9%** | 24.1% | 35.1% | `HIGH` |
| **Civil Administration & Public Records** | Public Sector & Governance | 5.2 turns | `Synthesize & Formulate` | **48.6%** | 14.5% | 51.4% | `MODERATE` |
| **Healthcare Navigation & Clinical Intake** | Healthcare & Life Sciences | 6.8 turns | `Triage & Transcribe` | **41.3%** | 8.2% | 58.7% | `MODERATE` |

---

## 2. The 6-Tier AI Delegation Ladder (D0 to D5):
| Stage | Stage Name | Workload Share % | 90d Delta | Median Turns | Unreviewed Diffs % | Primary Interface | Fragility Risk |
| :---: | :--- | :---: | :---: | :---: | :---: | :--- | :---: |
| **D0** | Manual Baseline | **7.2%** | -18.4% | 1 | 0% | Manual IDE / Text Editor / Blank Canvas | `MINIMAL` |
| **D1** | Assisted Lookup | **14.8%** | -12.6% | 2.1 | 8.4% | Browser Web Chat UI (ChatGPT, Claude web, Gemini) | `LOW` |
| **D2** | Augmented Co-Pilot | **22.4%** | -4.2% | 4.8 | 24.6% | IDE Inline Autocomplete (GitHub Copilot, Supermaven) | `MODERATE` |
| **D3** | Supervised Delegation | **29.6%** | +14.8% | 8.6 | 31.8% | IDE Multi-File Agent (Cursor Composer, Windsurf, Copilot Workspace) | `ELEVATED` |
| **D4** | Autonomous Orchestration | **18.4%** | +42.5% | 16.4 | 44.2% | Terminal CLI Loop (Claude Code, Aider, Goose, Cline) | `ELEVATED` |
| **D5** | Ambient Autonomy | **7.6%** | +88.2% | 34 | 58.6% | Daemon Background Worker (Cron, Event-Driven Agent Cloud) | `CRITICAL` |

---

## 3. Semantic Prompt Intent Shifts (90-Day Longitudinal Trajectory):
| Intent Name | Category | Current Share % | 90d Shift | Tokens (In / Out) | Repeat Query Loop % | Heuristic Satisfaction |
| :--- | :--- | :---: | :---: | :---: | :---: | :---: |
| **Learn & Understand** | Cognitive Knowledge | **11.2%** | -36% | 320 / 640 | 22.4% | 88.2/100 |
| **Search & Retrieve** | Information Retrieval | **14.6%** | -39.7% | 180 / 420 | 18.6% | 84.5/100 |
| **Decide & Compare** | Strategic Evaluation | **9.8%** | -14% | 580 / 890 | 31.2% | 82/100 |
| **Create & Draft** | Generative Production | **18.2%** | -20.2% | 420 / 1280 | 28.5% | 86.4/100 |
| **Solve & Debug** | Diagnostic Remediation | **16.8%** | +19.1% | 850 / 920 | 34.8% | 79.6/100 |
| **Delegate Task** | Autonomous Execution | **20.4%** | +148.8% | 1450 / 3600 | 46.2% | 91.2/100 |
| **Automate Pipeline** | Continuous Autonomy | **9%** | +400% | 2100 / 5800 | 52.4% | 93.5/100 |

---

## 4. Cross-Model Switching Flows & Workflow Churn Matrix:
| Source Workflow | Destination Workflow | Primary Catalyst Driver | 90d Churn % | Friction Index | Displacement Velocity |
| :--- | :--- | :--- | :---: | :---: | :---: |
| **undefined** | **undefined** | undefined | **undefined%** | undefined/100 | `undefined` |
| **undefined** | **undefined** | undefined | **undefined%** | undefined/100 | `undefined` |
| **undefined** | **undefined** | undefined | **undefined%** | undefined/100 | `undefined` |
| **undefined** | **undefined** | undefined | **undefined%** | undefined/100 | `undefined` |
| **undefined** | **undefined** | undefined | **undefined%** | undefined/100 | `undefined` |
| **undefined** | **undefined** | undefined | **undefined%** | undefined/100 | `undefined` |
| **undefined** | **undefined** | undefined | **undefined%** | undefined/100 | `undefined` |
| **undefined** | **undefined** | undefined | **undefined%** | undefined/100 | `undefined` |

---

## 5. Cognitive Reliance Radar & Automation Bias Index:
| Reliance Dimension | Workload Domain | Unreviewed Diffs % | Diff Review Time | Escalation Frequency | Automation Bias Risk |
| :--- | :--- | :---: | :---: | :---: | :---: |
| **Unreviewed Git Commits in CLI Loops** | Software Engineering | **44.2%** | 14s | 28.6% | `EXTREME` |
| **Phantom Package & Dependency Acceptance** | Software Engineering & IT | **38.5%** | 8s | 34.2% | `EXTREME` |
| **Secret & API Credential Leakage in Prompts** | Enterprise IT & DevOps | **29%** | 6s | 42% | `EXTREME` |
| **Scrutiny Fatigue Decay (Hour 1 vs Hour 4)** | Cross-Domain Knowledge Work | **52.8%** | 12s | 31.4% | `HIGH` |
| **Tautological & Mocked Unit Test Blindness** | Software Engineering | **41.6%** | 18s | 26.8% | `HIGH` |
| **Statutory Citation Rubber-Stamping** | Legal M&A & Regulatory | **22.4%** | 48s | 18.5% | `HIGH` |
| **Context Window Truncation Forgetting** | Architecture & Modeling | **36.2%** | 22s | 24% | `MODERATE` |
| **Silent Edge-Case Omission in Business Logic** | Financial & E-Commerce Systems | **33.5%** | 26s | 29.5% | `HIGH` |
| **Junior Engineer Mental Model Atrophy** | Software Engineering & Data Science | **48%** | 16s | 38% | `EXTREME` |
| **Multi-Agent Cascade Error Compounding** | Autonomous Multi-Agent Orchestration | **46.5%** | 20s | 36.4% | `EXTREME` |

---

## REST Endpoints & Machine Discovery:
- Cryptographic Proof: GET https://api.aki1k.com/v1/behaviour/verify
- Macro JSON Telemetry: GET https://api.aki1k.com/v1/behaviour/summary
- Workload Domains: GET https://api.aki1k.com/v1/behaviour/workloads
- Delegation Ladder: GET https://api.aki1k.com/v1/behaviour/delegation
- Prompt Intents: GET https://api.aki1k.com/v1/behaviour/intents
- Switching Flows: GET https://api.aki1k.com/v1/behaviour/switching
- Reliance Radar: GET https://api.aki1k.com/v1/behaviour/reliance
- Displacement Velocity: GET https://api.aki1k.com/v1/behaviour/displacement
- Official Web Hub: https://aki1k.com/behaviour
- Delegation Hub: https://aki1k.com/behaviour/delegation
- Intents Hub: https://aki1k.com/behaviour/intents
- Switching Hub: https://aki1k.com/behaviour/switching
- Reliance Radar Hub: https://aki1k.com/behaviour/reliance-radar
- Displacement Hub: https://aki1k.com/behaviour/displacement