Our Philosophy on Artificial Intelligence

Legal problems are often expensive to address because traditional attorney time is scarce and costly. Many people never get help — not because the law is unknowable, but because organizing facts, understanding options, and preparing documents takes more time and money than they can afford.

We believe artificial intelligence, used responsibly and under attorney supervision, can change that equation without sacrificing the judgment, empathy, and accountability that only a licensed human attorney can provide.

Our philosophy rests on three convictions:

1. AI is exceptionally good at helping people understand what the law is. Large language models can synthesize statutes, rules, and common legal concepts into plain English, help a person see the legal landscape around their situation, and surface issues they might not have known to ask about.
2. AI is exceptionally good at organizing facts into a useful file. Through structured intake, AI can help gather, categorize, and present a person's story — dates, parties, documents, open questions — in a Living File that makes attorney review faster and more focused.
3. A licensed attorney must always be in the loop. AI does not exercise legal judgment, does not form an attorney-client relationship on its own, and cannot replace the human considerations — context, empathy, strategy, ethics, and accountability — that legal representation requires. Every AI-assisted output that matters is reviewed by a licensed attorney before it is delivered to a client as approved.

Depending on which phase of the platform you use, AI may assist with:

Phase I uses AI to provide general legal information. It does not create an attorney-client relationship, is not protected by privilege, and is not legal advice. Do not share confidential or sensitive facts in Phase I.

Phase II uses AI within a signed representation agreement. AI assists intake, analysis, and drafting; a licensed attorney supervises the process and reviews documents before they are delivered as attorney-approved.

To be direct about limitations — because candor is an ethical obligation, not a marketing inconvenience:

• AI is not your attorney. It is a tool Crawford Law uses under attorney supervision. It does not hold a law license and cannot be held accountable in the ways a human lawyer can.
• AI can be wrong. Generative AI systems can produce incomplete, outdated, or fabricated information — including citations and legal conclusions that sound authoritative but are not (“hallucinations”). We do not ask you to trust AI output. We ask you to trust the attorney review process that follows it, while understanding that no process is perfect.
• AI does not guarantee outcomes. No result in any legal matter is guaranteed. AI-assisted analysis is a starting point for professional judgment, not a prediction of what a court, agency, or opposing party will do.
• AI does not replace human judgment on matters that require it. Decisions involving your safety, your children, your liberty, your livelihood, or your fundamental rights deserve human attention. We use AI to make attorney time more efficient and accessible — not to remove the attorney from decisions that require human wisdom.

This is not a product feature we chose for differentiation. It is an ethical requirement under the Texas Disciplinary Rules and Texas Ethics Opinion 705.

1. Supervision. A licensed attorney (Andrew Crawford, Esq.) supervises how AI is used in client matters and is responsible for the work product delivered.
2. Review before delivery. AI-generated documents are labeled as pre-review and watermarked until a licensed attorney reviews and approves them. Pre-review drafts must not be filed, signed, or relied upon.
3. Independent verification. We do not blindly rely on AI research, citations, or legal conclusions. Attorney review includes checking accuracy, completeness, and fitness for the client's specific situation.
4. Competence. We maintain a current, reasonable understanding of the AI tools we use, their capabilities, and their risks — as Opinion 705 requires.
5. Communication. We tell clients that AI is used, how it is used, and what its limitations are — through this statement, onboarding consent, and the AI Consent document.

Confidentiality is non-negotiable. Under Tex. Disciplinary R. Prof. Conduct 1.05, we treat Phase II client information as confidential.

• Single AI provider. We use Anthropic, PBC as our AI inference provider, deliberately limiting vendor complexity so we can enforce strong data-handling terms.
• No training on your content. We do not permit your communications to be used to train AI models.
• Zero-data-retention (ZDR). We are implementing ZDR processing with our AI provider before public launch, so that prompt and output content is not retained by the provider.
• Vetted service providers. Cloud hosting, payments, and email are handled by service providers under contractual confidentiality and data-protection obligations.
• Informed consent. Before Phase II services begin, you sign an AI Consent document that describes third-party processing and your right to revoke consent.

We use AI because we believe more people deserve to understand their legal situation and take meaningful steps — not because we believe legal services should be automated away.

Traditional full-scope representation is the right answer for many matters. It is also out of reach for many people. Our limited-scope, AI-assisted model is designed to help people who might otherwise do nothing: understand the law, organize their facts, receive attorney-reviewed documents, and decide — with real information — whether and how to proceed.

Lower cost does not mean lower ethical standards. The same rules of professional conduct apply whether an attorney spends ten hours on a matter or uses AI to make those ten hours more productive.

We are transparent about how AI usage affects fees:

• The Phase II subscription and automatic usage top-ups are described in your Representation Agreement and the Billing & Refund Disclosure.
• We do not charge hourly rates for time saved by AI in ways that would be misleading under Opinion 705 and Rule 1.04 (fees).
• AI usage has a real cost; that cost is reflected in our subscription and metered top-up structure, not hidden.

• Read before you sign. This philosophy statement, the AI Consent, and the Representation Agreement are available before you subscribe.
• Revoke consent. You may revoke AI consent at any time by contacting the Firm in writing. Because AI assistance is integral to how the platform works, revocation generally means we can no longer provide platform services.
• Raise concerns. If any AI-assisted output seems wrong, incomplete, or inappropriate, contact us before acting on it.
• Export your data. You may export your documents and Living File during your engagement.

We commit to the following practices, aligned with Texas Ethics Opinion 705 and the Texas Disciplinary Rules:

If you have questions about how we use AI, contact:

Crawford Law PLLC
Andrew Crawford, Esq. · Texas Bar #24148908
www.instant-attorney.com