The intake assistant
How the intake assistant works
The assistant on this site is an artificial intelligence (AI) system. It is not a person, and it is not a lawyer.
What it is
The intake assistant runs on Claude, a large language model developed by Anthropic, PBC. It is provided through the CITE8 intake platform, which Schwartz & Associates, PC licenses from CITE8, Inc., a legal technology company affiliated with the firm's principal. The assistant follows written instructions set by the firm and relies on a reference list of pending securities class actions and mass tort litigation that the firm compiles from public court filings, published notices, government records, and news reports.
What it can do
- Answer general questions about securities class actions, lead plaintiff deadlines, mass torts, and multidistrict litigation.
- Explain the firm's published flat fees for contract review, negotiation, and drafting.
- Tell you what the firm's reference list says about a pending case or litigation, such as a reported class period or deadline.
- Ask a few questions so the firm can check for conflicts and review your matter, and prepare a short form for you to review, correct, and send.
What it cannot do
- Give legal advice about your situation, tell you whether you have a claim, or predict an outcome or an amount.
- Decide whether the firm will take your matter. Only a lawyer can make that decision.
- Create an attorney-client relationship, or pause any deadline that applies to your claim.
- Accept documents. Please do not send account numbers, Social Security numbers, or medical records.
Voice input
On browsers that support it, you can speak instead of typing. Your browser converts your speech to text using its own speech service (for example, Google in Chrome or Apple in Safari), and the text appears in the message box for you to review before you send it. The site does not record, receive, or keep audio; the firm receives only the text you send. Speech-to-text can mishear words, so check names and dates before you send.
You can also have a spoken conversation with the assistant by choosing Talk with the assistant. Your speech is streamed directly from your browser to OpenAI, L.L.C. GPT-Live, OpenAI's voice model, listens and replies in a synthetic voice, and GPT-5.6 Terra, an OpenAI language model, applies instructions based on the same intake rules as the text chat. The firm does not record or keep audio. OpenAI processes the conversation under its API terms, which state that API data is not used to train its models by default, and it may retain data for a limited period as described in its API data policies. Permission requests, forms, and consents still appear on your screen and are completed there.
Health information
If your matter involves an injury or illness, the assistant asks your permission before asking about your health. You can decline and still reach a person at the firm. See the Consumer Health Data Privacy Policy.
Sensitive matters
Matters involving sexual assault or abuse, children, or a death are handled by a person at the firm. The assistant will not ask you about what happened. It will give you a way to reach the firm directly.
Accuracy
The assistant can make mistakes. Case information comes from public sources and can change, and court records control over anything the assistant says. Deadlines are compiled from published notices and dockets; confirm any deadline on the court docket before relying on it.
Supervision and records
The firm sets and reviews the assistant's instructions and may keep a record of the assistant's responses to supervise their accuracy. When you send a form, the firm receives the form together with the conversation that led to it; if you declined permission to collect health information, your own messages are left out. The firm does not otherwise store conversations that you do not send.
Your information and the AI provider
To operate the assistant, your messages are sent to Anthropic for processing. Under Anthropic's commercial terms, this information is not used to train its models. Anthropic states that it deletes inputs and outputs within 30 days by default, subject to limited exceptions such as legal requirements and enforcement of its usage policies. Numbers that look like Social Security, card, or account numbers are removed from typed messages before they are sent. Spoken conversations are processed by OpenAI, which states that data sent through its API is not used to train its models by default; the assistant asks you not to say such numbers aloud. See the Privacy Notice.
Talk to a person
You can call the firm at (918) 238-6445 at any time, or use the Talk to a person button in the chat.