# AI for Customer Support, Success & Experience URL: https://justuju.org/role/support/ For your role Support is where AI agents are most established. They answer from your help content, take actions such as refunds within limits, and work in chat, email and voice. Human agents move to the hard cases, with AI drafting beside them. ## How AI is changing this work - First contact is increasingly an AI agent that resolves the issue, not a form that deflects it. - Actions such as refunds, bookings and account changes are done by the AI under written rules. - Voice agents now hold natural phone conversations, replacing menu trees. ## Key layers, most important first ### 1. Layer 9 · Applications & Interfaces URL: https://justuju.org/layer/applications-interfaces/ Why it matters: The customer meets your AI here, in chat, email or on the phone. Tone, speed and the path to a person are the experience. What to do here: - AI agent on every channel: One agent answers in chat, email and voice from the same knowledge and rules. (see https://justuju.org/concept/chat-assistants/) - Voice that listens: Spoken agents handle interruptions and accents, and replace phone menus. (see https://justuju.org/concept/voice-multimodal/) - Assist for human agents: Suggested replies, summaries and next steps appear beside the ticket. (see https://justuju.org/concept/copilots/) Examples as of 2026-11-11: Fin (Intercom) (https://fin.ai/); Zendesk AI agents (https://www.zendesk.com/blog/zendesk-insights/innovation/relate-2025-resolution-platform-ai-agents/); Customer Agent (HubSpot) (https://www.hubspot.com/company-news/hubspots-customer-agent-and-prospecting-agent-now-you-pay-when-the-task-is-complete) ### 2. Layer 8 · Agents URL: https://justuju.org/layer/agents/ Why it matters: Answering is not resolving. Resolution needs an agent that can look up the order, apply the policy and act, within limits you set. What to do here: - Procedures in plain language: Write each workflow, such as refunds under 30 days, as steps the agent follows and engineers can test. (see https://justuju.org/concept/agent-loop/) - Limits on actions: Cap amounts, require confirmation for account changes, and log every action. (see https://justuju.org/concept/agent-permissions/) - Clean handoff: Escalate on request, on repeated failure or on high value, passing the full history to the person. (see https://justuju.org/concept/agent-loop/) Examples as of 2026-11-11: Sierra (https://sierra.ai/product/agent-studio); Decagon (https://decagon.ai/blog/ai-customer-service-agent-capabilities) ### 3. Layer 7 · Orchestration & Tooling URL: https://justuju.org/layer/orchestration-tooling/ Why it matters: The agent is only as right as your help content and as safe as its guardrails. Customers will try to talk it into things. What to do here: - Answer from your knowledge base: Ground every answer in current help articles and policies, and fix the article when the answer is wrong. (see https://justuju.org/concept/rag/) - Test with real tickets: Replay past conversations before launch and after every change, and score the results. (see https://justuju.org/concept/guardrails-evals/) - Expect manipulation: Customers and web content can try to override its rules. Screen inputs and keep limits outside the model. (see https://justuju.org/concept/prompt-injection/) Examples as of 2026-11-11: Glean (https://www.glean.com/platform); Prompt Shields (Azure AI Content Safety) (https://learn.microsoft.com/en-us/azure/ai-services/content-safety/whats-new) ### 4. Layer 6 · Inference & Optimization URL: https://justuju.org/layer/inference-optimization/ Why it matters: Support runs at volume and in real time. Cost per conversation and the pause before a spoken reply are both decided by how the model is served. What to do here: - Speed for voice: A phone conversation needs replies in well under a second; choose models and hosting for that. (see https://justuju.org/concept/model-apis/) - Small model first: Send simple questions to a small fast model and only hard ones to a large one. (see https://justuju.org/concept/model-apis/) - Cache the fixed part: Policies and instructions repeat in every conversation; caching them cuts cost. (see https://justuju.org/concept/batching-caching/) Examples as of 2026-11-11: OpenAI Realtime API (https://developers.openai.com/api/docs/guides/realtime); ElevenLabs Agents (https://elevenlabs.io/agents); Prompt caching (Claude API) (https://platform.claude.com/docs/en/build-with-claude/prompt-caching) ## High-leverage use cases - Resolve common questions: Order status, how-to and policy questions answered at once, with a source. - Take routine actions: Refunds, returns, bookings and address changes within written limits. - Agent assist: Draft replies and summaries for human agents, which helps new staff most. - Voice front door: A spoken agent that understands the request and resolves or routes it. - Insight from conversations: Themes and product issues surfaced from all contacts, not a sample. ## Start here: the next 30 to 60 days - Week 1 to 2: Find your ten most common questions and check the help articles for each are current and clear. - Week 3 to 4: Launch an AI agent on one channel for those topics, with an easy route to a person. - Day 30 to 60: Review 50 AI conversations a week. Fix content first, rules second. - Day 30 to 60: Add one action, such as order lookup or a capped refund, with logging and limits. ## Risks and pitfalls - You own what it says: A Canadian tribunal held an airline responsible for wrong information its website chatbot gave a customer. Source: https://www.mccarthy.ca/en/insights/blogs/techlex/moffatt-v-air-canada-misrepresentation-ai-chatbot - Telling customers it is AI: In the EU, people must be told when they are talking to an AI unless it is obvious. Source: https://eur-lex.europa.eu/eli/reg/2024/1689/oj - No way out: Customers trapped with a bot leave. Always offer a person, and escalate on frustration. - Talked into exceptions: Customers will test whether the agent can be argued into refunds or discounts. Enforce limits in the system, not in the prompt. Source: https://genai.owasp.org/llm-top-10/ - Measuring deflection, not resolution: A closed chat is not a solved problem. Track repeat contacts and satisfaction. ## Sources - Fin (Intercom) (fin.ai): https://fin.ai/ - Zendesk AI agents (zendesk.com): https://www.zendesk.com/blog/zendesk-insights/innovation/relate-2025-resolution-platform-ai-agents/ - Customer Agent (HubSpot) (hubspot.com): https://www.hubspot.com/company-news/hubspots-customer-agent-and-prospecting-agent-now-you-pay-when-the-task-is-complete - Sierra (sierra.ai): https://sierra.ai/product/agent-studio - Decagon (decagon.ai): https://decagon.ai/blog/ai-customer-service-agent-capabilities - Glean (glean.com): https://www.glean.com/platform - Prompt Shields (Azure AI Content Safety) (learn.microsoft.com): https://learn.microsoft.com/en-us/azure/ai-services/content-safety/whats-new - OpenAI Realtime API (developers.openai.com): https://developers.openai.com/api/docs/guides/realtime - ElevenLabs Agents (elevenlabs.io): https://elevenlabs.io/agents - Prompt caching (Claude API) (platform.claude.com): https://platform.claude.com/docs/en/build-with-claude/prompt-caching - Moffatt v. Air Canada: a misrepresentation by an AI chatbot (McCarthy Tétrault, on 2024 BCCRT 149): https://www.mccarthy.ca/en/insights/blogs/techlex/moffatt-v-air-canada-misrepresentation-ai-chatbot - Regulation (EU) 2024/1689 (the AI Act): https://eur-lex.europa.eu/eli/reg/2024/1689/oj - OWASP Top 10 for LLM Applications: https://genai.owasp.org/llm-top-10/ - Generative AI at Work (Brynjolfsson, Li, Raymond): https://arxiv.org/abs/2304.11771 --- Justuju · https://justuju.org/ · Content CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote with a link to the page.