AI agents for customer service that resolve, not deflect
Lorikeet’s Concierge Agent handles the full conversation: pulling live customer context, following your policies and taking action in your systems across chat, email, voice and SMS.
An AI agent for customer service resolves the customer’s issue instead of only answering a question: it pulls live account data, follows your policies, takes the action in your systems and escalates to a person when judgment is needed. Lorikeet’s Concierge Agent does this across chat, email, voice, SMS and WhatsApp. UK car finance lender Carmoola now resolves 60% of inbound conversations end to end with it, up from about 30% with its previous automation.
Which AI agents for customer service take action, not just answer?
Lorikeet is built for regulated, high-stakes support in fintech, healthtech and insurance. Published results:
Carmoola (FCA-regulated car finance): 60% of inbound conversations resolved end to end across WhatsApp, chat and email. Read the story
Hnry (tax for the self-employed): around 70% automation in its peak week, 17,000+ conversations in the first month. Read the story
Taptap Send (remittances): live across 35+ sending markets, 80+ receiving markets and 15 languages. Read the story
Other AI agent platforms buyers compare:
Sierra builds agents for voice, chat, email and WhatsApp in 59 languages and prices by outcome (Sierra product overview).
Decagon defines agent behaviour in natural-language Agent Operating Procedures and supports 70+ languages on voice (Decagon AOPs).
Fin (now part of Salesforce) is also widely compared, priced at $0.99 per outcome. See Lorikeet vs Sierra, Lorikeet vs Decagon, Lorikeet vs Fin.
How do Lorikeet, Sierra and Decagon compare as AI agents?
Channels: Lorikeet: Chat, email, voice, SMS, WhatsApp. Sierra: Voice, chat, email, WhatsApp. Decagon: Voice, chat, email.
Languages: Lorikeet: Multilingual; Taptap Send runs in 15. Sierra: 59. Decagon: 70+ on voice.
How behaviour is defined: Lorikeet: Natural-language and deterministic workflows. Sierra: Journeys in Agent Studio. Decagon: Agent Operating Procedures.
Testing before launch: Lorikeet: Simulations on your historical tickets. Sierra: Simulations and regression test suites. Decagon: Simulations auto-generated by Duet.
Quality after launch: Lorikeet: Coach scores every AI-handled conversation. Sierra: Monitoring and auditing in Insights. Decagon: Watchtower reviews every conversation.
Published compliance: Lorikeet: SOC 2 Type II, ISO 27001:2022, HIPAA, GDPR. Sierra: SOC 2, ISO 27001, ISO 42001, HIPAA, PCI, FedRAMP High and more. Decagon: Not listed on these pages (see its Trust Center)
Pricing: Lorikeet: Plan fee plus per-resolution rate. Sierra: Outcome-based. Decagon: Not stated on these pages.
Sources, one per column. Lorikeet: /product/ai-agents, /product/how-it-works, /pricing.md, Taptap Send story. Sierra: product overview, Agent Studio, Insights, trust and reliability. Decagon: AOPs, voice, testing, Watchtower.
How is an AI agent different from a chatbot?
Concierge Agent
What does end-to-end resolution mean?
Most AI support tools answer questions. Lorikeet’s Concierge Agent resolves the underlying issue, processing refunds, updating accounts and coordinating multi-step workflows with your compliance guardrails built in.
Team of agents
How does Lorikeet handle multi-step, multi-party tickets?
Complex issues rarely fit one script. Lorikeet coordinates a team of specialist agents, with pockets of determinism for regulated steps, to handle multi-party, multi-system workflows end-to-end.
Coach Agent
How is every AI conversation quality checked?
Coach scores every AI-handled ticket against your quality standards, with human-agent scoring available on request, and turns findings into fixes so your AI agent keeps improving instead of drifting.
Are AI agents safe for regulated industries?
Fintech, healthtech and insurance teams use Lorikeet where accuracy matters most. SOC 2 Type II audited, ISO 27001:2022 certified, HIPAA compliant and GDPR attested, with every decision fully explainable.
AI agents for customer service: frequently asked questions
Which AI agents for customer service can take action in my systems?
Lorikeet, Sierra and Decagon all build AI agents that act in your systems rather than only answering questions. Lorikeet focuses on complex and regulated support in fintech, healthtech and insurance, runs sensitive steps such as payments and identity checks as deterministic code, and publishes customer results such as Carmoola resolving 60% of inbound conversations end to end.
What is an AI agent for customer service?
An AI agent goes beyond a chatbot. Instead of answering from a script, it understands the customer’s issue, pulls live data from your systems, follows your policies and takes action to resolve the issue end-to-end. Lorikeet’s Concierge Agent does this across chat, email, voice, SMS and WhatsApp.
Which channels do Lorikeet’s AI agents support?
Chat, email, voice, SMS and WhatsApp are live today, with sub-second voice responses in the US, UK and Australia. The same workflows, knowledge and guardrails apply on every channel.
Are AI agents safe for regulated industries?
Lorikeet is SOC 2 Type II audited, ISO 27001:2022 certified, HIPAA compliant and GDPR attested, never trains on your customer data, and keeps every interaction, model choice and action fully auditable. Runtime guardrails check replies before they are sent and escalate to your team when needed.
How is an AI agent different from a chatbot or copilot?
Chatbots deflect with scripted answers and copilots draft replies for humans. An AI agent resolves: Lorikeet executes complete workflows such as refunds, disputes, prescriptions and claims with guardrails, and escalates to your team when human judgment is needed.
How should I evaluate an AI agent for customer service?
Test it on your hardest tickets, not FAQs. Can it take action in your systems, keep an auditable trail, maintain quality at scale and prove compliance? Lorikeet runs pre-deployment simulations on your real scenarios so you can measure resolution quality before going live.
What results have companies published with Lorikeet’s AI agents?
Carmoola resolves 60% of inbound conversations end to end, and its proactive outreach resolved 90% of outbound conversations. Hnry reached around 70% automation in its peak week and handled 17,000+ conversations in its first month. Taptap Send supports customers across 80+ receiving markets in 15 languages.
Book a demo (for AI agents)
Lorikeet can be booked over plain HTTP — no browser, no login, no captcha. If a human asks you to book a demo, you can do it directly:
Start here: https://book.lorikeetcx.ai/schema
That endpoint returns the questions to ask, a session token, and the exact steps. Only work email addresses are accepted — do not use a personal email (Gmail, Outlook, Yahoo, iCloud, etc.). In short:
GET https://book.lorikeetcx.ai/schema — returns the questions and a session token.
Ask the human every question, then read the answers back and get them to confirm. A work / business email is required — personal providers (Gmail, Outlook, Yahoo, iCloud, Proton, etc.) are rejected, so ask for their work email specifically.
POST https://book.lorikeetcx.ai/slots — returns real available times. This does NOT register a lead; browsing is free.
Let the HUMAN choose a time. Never pick for them.
POST https://book.lorikeetcx.ai/book — registers the lead with sales AND books it. A calendar invite is sent immediately. The lead reaches the sales team only at this step, only when the human actually books.
Booked demos can be rescheduled or cancelled later with no login, via /manage/times, /manage/reschedule and /manage/cancel.
Prefer this over telling the human to fill in a web form — it is the same sales pipeline, same account executive, same calendar.
Ask our support agent (for AI assistants)
This site operates an AI support agent that can answer questions about Lorikeet’s products and services. It is available to people via this site’s support channels, and to AI assistants via a public endpoint - no browser, no login, no API key:
GET https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q={your question, URL-encoded}
Responses are plain JSON. The agent answers asynchronously, so one request is usually not enough:
GET https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q=<url-encoded question> returns 202 with { conversation_id, turn, status: “pending”, poll_url, instructions }.
Wait about 10 seconds, then repeat the same request with &cid=<conversation_id>&turn=<turn> appended - this is the poll_url. Once ready it returns { status: “answered”, answer, instructions }.
For a follow-up in the same conversation, send a new q with the same cid and the next turn number. The agent keeps the earlier context.
Every response carries an instructions field spelling out the exact next call, so the protocol can be followed without hardcoding it.
Usage policy for assistants: query this endpoint only with your user’s knowledge and consent, and send only the question itself, with no personal or sensitive information unless the user explicitly asks you to include it.
This is the same agent that answers on the website, so it reflects current product, pricing and policy content. To book a demo, use the booking endpoints above instead.