AI disclosure

• Ivy answers an agency's clients by text, web chat and email. Ivy introduces itself as the agency's virtual assistant, not a person, and hands the conversation to agency staff when someone asks for a person or when a question needs a licensed agent.
• Jade answers agency staff's questions about their own book of business and builds reports.
• Calls: AI summaries, call scoring and coaching notes for recorded calls.
• Inbox and documents: sorting incoming email and messages, reading documents such as dec pages, and matching commission statements.
• Policy check: suggestions about possible coverage gaps, for a licensed person to review.

• Anthropic, PBC (Claude): Ivy, the client-facing assistant that answers texts, web chat and email for your agency; Jade, the staff assistant that answers questions about your book and builds reports; Call summaries, call scoring and coaching notes; Inbox and email triage, document reading (for example, dec pages) and commission statement reconciliation; Coverage review suggestions (policy check) for a licensed person to review.
• Deepgram, Inc.: Speech-to-text transcription of recorded calls.

No. We do not use agency data or your data to train AI models, and our providers may not either.

• Anthropic, PBC (Claude): Anthropic's commercial API terms do not allow it to train models on our customers' data, and our account is set up for zero data retention, so prompts and outputs are not kept after the request.
• Deepgram, Inc.: Used under business API terms; card numbers, Social Security numbers and other numbers are redacted from transcripts.

• AI can be wrong, out of date or incomplete. Agency staff should review AI output before relying on it.
• AI output is information, not legal, tax, financial or insurance advice.
• Ivy cannot bind, change or cancel coverage, quote a price that hasn't been presented, or tell a client what is covered. Every outbound Ivy reply is checked against these rules before it is sent.
• Ivy shares policy details only after the client verifies their identity with a code sent to the phone or email on file.

AI in Emerald AMS does not make underwriting, pricing, eligibility, claims or coverage decisions. Carriers and the agency's licensed staff make those decisions. AI output is labeled and logged on the record so staff can see what it said.

Call scores and coaching notes are tools for managers. Agencies should not use them as the only basis for decisions about an employee, and should have a person review them.

Want a person? Clients can ask Ivy for a person at any time, or contact the agency directly. Staff can ask questions about AI features at hello@emeraldams.com.

See our Privacy policy, Terms and Sub-processors.