---
title: "Build an escalation matrix for a client-facing legal tool"
url: "https://ailawframework.com/obligations/client-facing-and-access-to-justice-ai-tools/build-an-escalation-matrix-for-a-client-facing-tool/united-states"
type: "task_jurisdiction_version"
site: "AI Law Framework"
id: "A43-A20260923-1-US"
practice_area: "Client-facing & access-to-justice AI tools"
tier: "yellow"
lawyer_review: "required"
jurisdictions: ["US"]
last_reviewed: "2026-10-22"
---

# Build an escalation matrix for a client-facing legal tool

> Turn the firm's tool description into a draft matrix of who is told, and within what time, when the tool gives advice, takes a step, or mishandles client information.

Canonical page: https://ailawframework.com/obligations/client-facing-and-access-to-justice-ai-tools/build-an-escalation-matrix-for-a-client-facing-tool/united-states  
Last reviewed: 2026-10-22

Site index for agents: https://ailawframework.com/llms.txt

## Key facts

| Fact | Value |
| --- | --- |
| Practice area | [Client-facing & access-to-justice AI tools](https://ailawframework.com/obligations/client-facing-and-access-to-justice-ai-tools.md) |
| Human involvement | Yellow: Lawyer and AI draft together |
| How AI is used | Turn the firm's tool description into a draft matrix of who is told, and within what time, when the tool gives advice, takes a step, or mishandles client information. |
| What the lawyer must review | You set the thresholds. ABA Formal Opinion 512: the tool cannot offer legal advice, negotiate or file without your judgment. Model Rule 5.5(a) forbids assisting unauthorised practice. California's 2026 guidance: an agentic system must not make substantive legal determinations without meaningful lawyer supervision. Model Rules 5.1 and 5.3 require a real supervision policy, not a chart the tool wrote. |
| Written for | United States |
| Applies to | [Build an escalation matrix for a client-facing tool](https://ailawframework.com/obligations/client-facing-and-access-to-justice-ai-tools/build-an-escalation-matrix-for-a-client-facing-tool.md) |

## Checking plan

- What the AI does: Turn the firm's tool description into a draft matrix of who is told, and within what time, when the tool gives advice, takes a step, or mishandles client information.
- Basis: Checked against the linked sources on 22 October 2026. Confirm the current instrument before reliance.

## Notes

- Check the admitting state's version of Rules 5.1, 5.3 and 5.5, and any court order on client-facing tools.
- A public chatbot that answers a user's legal question can be unauthorised practice even if the matrix says a lawyer will follow up.

## Sources

- [ABA Formal Opinion 512, Generative Artificial Intelligence Tools (29 July 2024)](https://www.americanbar.org/content/dam/aba/administrative/professional_responsibility/ethics-opinions/aba-formal-opinion-512.pdf) (checked 2026-10-22)
- [State Bar of California, Practical Guidance for the Use of Generative Artificial Intelligence in the Practice of Law (2026)](https://www.calbar.ca.gov/Portals/0/documents/ethics/Generative-AI-Practical-Guidance.pdf) (checked 2026-10-22)
- [ABA Model Rule 5.5, Unauthorized Practice of Law](https://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_professional_conduct/rule_5_5_unauthorized_practice_of_law_multijurisdictional_practice_of_law/) (checked 2026-10-22)

## Forms and precedents for this task

| Form or precedent | Kind | AI level | Summary |
| --- | --- | --- | --- |
| [Client consent to input information into a generative AI tool (United States)](https://ailawframework.com/precedents/client-facing-and-access-to-justice-ai-tools/us-client-consent-input-to-generative-ai.md) | Letter | Yellow | A tool-specific consent for putting information relating to a representation into a named generative AI tool, with the risk particulars Formal Opinion 512 says a boilerplate engagement sentence does… |
| [No-advice scope and blocked-topics schedule for a client-facing assistant](https://ailawframework.com/precedents/client-facing-and-access-to-justice-ai-tools/no-advice-scope-schedule.md) | Policy | Red | Schedule the supervising lawyer adopts for a client-facing assistant: topics it may answer with general information, topics it must refuse, and the exact handoff it must give for urgency, safety and… |
| [No-advice notice for a client-facing tool (United States)](https://ailawframework.com/precedents/client-facing-and-access-to-justice-ai-tools/us-client-facing-tool-no-advice-notice.md) | Notice | Yellow | A fixed notice that a client-facing tool is not a lawyer, does not give legal advice, and does not create a client-lawyer relationship. The firm approves the words once. The tool does not vary them. |
| [Note escalating a client-facing tool transcript to a lawyer (United States)](https://ailawframework.com/precedents/client-facing-and-access-to-justice-ai-tools/us-chatbot-escalation-note.md) | Memo | Yellow | An internal note that lifts a transcript out of a client-facing tool and records why a lawyer must take the next step. It does not answer the person. |
| [Periodic supervision minute for a client-facing AI tool](https://ailawframework.com/precedents/client-facing-and-access-to-justice-ai-tools/client-facing-tool-supervision-minute.md) | Memo | Red | Minute in which the supervising lawyer records, for a set period, what the client-facing tool did, what the sample review and incidents showed, and whether it stays on, changes, or is switched off. |

Signed-in accounts can read the full text of each one.

*Practice guidance for planning and review. Confirm current primary sources and the requirements of the actual matter. This is general information, not legal advice.*