The Aide Glossary:
Agentic AI in customer experience
Definitions for the vocabulary of agentic AI in customer service and experience: what the terms mean, how they differ, terms they are commonly confused with, and why those differences matter.AI and Agentic fundamentals
What agentic AI actually is, how it differs from generative and conversational AI, and the platform concepts underneath.- Agent harness
- Agent memory
- Agent orchestration
- Agent OS (agent operating system)
- Agentic AI
- Agentic AI for customer experience
- Agentic AI maturity model
- Agentic AI platform
- Agentic AI vs conversational AI
- Agentic AI vs generative AI
- Agentic Customer Experience (ACX)
- Agentic process automation
- Agentic RAG
- Agentic workflow
- AI agent for customer service
- AI agent platform
- AI agent vs AI assistant
- AI agent vs chatbot
- AI agents for insurance
- AI agents vs copilots
- AI concierge
- AI customer service agent
- AI customer service for e-commerce
- AI customer service for logistics
- AI customer service for SaaS
- Autonomous customer support
- Chatbot automation
- Context engineering
- Context window
- Conversational AI
- Conversational AI design
- Conversational AI in banking
- Conversational AI in healthcare
- Customer experience automation
- E-commerce customer experience (ecommerce CX)
- Enterprise chatbot
- Enterprise conversational AI
- Fine-tuning
- Foundation model
- Intelligent virtual agent (IVA)
- Knowledge base ceiling
- Large language model (LLM)
- Level of autonomy (autonomy spectrum) in AI
- LLMOps
- Model Context Protocol (MCP)
- Multi-agent system
- Multilingual conversational AI
- Reasoning engine
- System prompt
- Token in AI
- Tool calling (function calling)
- Virtual assistant
Intents
An important primitive the whole discipline rests on: classification, taxonomies, sub-intents, drift, and the difference between an intent and a topic.- AI triage
- Auto-Tagging
- Contact driver
- Customer intent
- Customer Intent Engine
- Intent classification
- Intent Classification vs Keyword Routing
- Intent classifier
- Intent confidence threshold
- Intent Coverage Rate
- Intent detection
- Intent drift
- Intent Map
- Intent recognition
- Intent taxonomy
- Intent vs entity
- Intent vs topic
- Intent-based routing
- Intent-first architecture
- Long-Tail Intents
- Multi-Intent Detection
- Named-Entity Recognition (NER)
- Natural Language Processing (NLP)
- Natural language understanding (NLU)
- Slot Filling
- Sub-intent
- Utterance
Procedures and Deployment
How automation ships safely: ASOPs, testing before deploy, staged rollouts, and human-in-the-loop patterns.- Agent operating procedure (AOP)
- Agentic SOP (ASOP)
- AI agent action
- AI agent workflow
- Business logic for AI agents
- CX workflow automation
- Decision tree vs agentic procedure
- Deterministic vs probabilistic agent execution
- Intent-scoped deployment
- Natural-language workflow authoring
- Resolution playbook
- SOP for AI agents
Trust and Governance
Guardrails, hallucinations, grounding, explainability, action traces, and what it takes to put AI in front of customers without holding your breath.- Action trace
- Agent simulation testing
- Agent Simulator
- AI agent evaluation
- AI agent sandbox
- AI audit log
- AI escalation policy
- AI explainability
- AI governance for customer service
- AI grounding
- AI guardrails
- AI hallucination
- AI hallucination detector
- AI jailbreaking
- AI observability
- AI quality assurance
- AI red teaming
- AI risk scoring
- Confidence score
- Data privacy in AI customer service
- Guardrail vs guideline
- Hallucination guardrails
- HIPAA-compliant AI
- Human-in-the-loop
- Input and output guardrails
- LLM-as-a-judge
- PII redaction
- Prompt injection
- Regression testing for AI agents
- Responsible AI
- Retrieval-augmented generation (RAG)
- Staged rollout for AI agents
- Test-before-deploy
- Trust-engineered AI
- Zero data retention
Metrics and Operations
Resolution rate, deflection, containment, CSAT, NPS, handle time, and how to read them honestly.- AI coverage rate
- AI resolution rate
- Average handle time (AHT)
- Average resolution time (ART)
- Average speed of answer (ASA)
- Containment rate
- CSAT vs NPS vs CES
- Customer Effort Score (CES)
- Customer health score
- Customer satisfaction score (CSAT)
- Deflection rate
- Deflection rate vs resolution rate
- Escalation rate
- First contact resolution (FCR)
- Good NPS score
- Net Promoter Score (NPS)
- NPS detractor
- QA score
- Quality assurance in customer service
- Reopen rate
- Resolution rate
- Sentiment analysis (CX)
- SLA compliance
- Ticket volume
- Time to first response (FRT)
- Time to first value (TTFV)
- True resolution
- Why is deflection rate an incomplete metric
Helpdesk and contact center
The ground the work happens on: ticketing, routing, omnichannel, automation inside the helpdesk you already run.- After-call work
- Agent assist
- AI copilot for support
- Asynchronous messaging
- BPO customer service
- Canned response
- CCaaS
- Contact center automation
- Conversation analytics
- Escalation in customer support
- Helpdesk
- Helpdesk automation
- Human agent
- Knowledge base
- Knowledge management
- Omnichannel contact center
- Omnichannel customer support
- Omnichannel routing
- Self-service support
- Service-level agreement
- Support queue
- Ticket routing
- Ticketing system
- Tier 1 vs tier 2 support
- Workforce optimization
E-commerce CX
Orders, returns, WISMO, and the language of D2C support: the intents that fill a store’s queue and what resolving them actually takes.- Address change request
- AI order management
- BOPIS
- Chargeback response
- Conversational commerce
- Delivery exception
- Ecommerce chatbot
- Order cancellation automation
- Order modification
- Peak season customer service
- Post-purchase experience
- Pre-sale support
- Retail chatbot
- Returns automation
- RMA
- Subscription support
- WISMO
- WISMO automation
B2B SaaS support
Customer operations, technical support, and the tickets beyond the knowledge base: billing, bugs, APIs, and the accounts behind them.Financial services CX
Regulated conversations: disputes, hardship, payments, and what compliant automation means when every message is on the record.- AI for debt collection
- Banking chatbot
- Cease and desist request
- Collections automation
- Communication preference request
- Conversational AI in insurance
- Conversational banking
- Debt validation letter
- FDCPA compliant communication
- First-party vs third-party collections
- Hardship support
- Insurance chatbot
- KYC support
- Loan servicing support
- Net collections
- Payment plan automation
- Provisional credit
- Reg E error resolution
- Regulation F
- Transaction dispute
Logistics and freight CX
The freight inbox, term by term: status, quotes, documents, disputes, and the operational language of brokers, forwarders, and 3PLs.- Accessorial charges
- Agent based brokerage
- Bill of lading request
- Carrier scorecard
- Check call
- Customs entry status
- Delivery order release
- Detention and demurrage dispute
- Detention vs demurrage
- Dwell time
- EDI 214
- Exception management in freight
- Freight claims process
- Load status update
- OS&D
- Proof of delivery (POD) automation
- Quote to book rate
- Speed to quote
- TONU
- Track and trace
- WISMO in freight
All terms A–Z
- Accessorial charges
- Account based support
- Action trace
- Activation support
- Address change request
- After-call work
- Agent assist
- Agent based brokerage
- Agent harness
- Agent memory
- Agent operating procedure (AOP)
- Agent orchestration
- Agent OS (agent operating system)
- Agent simulation testing
- Agent Simulator
- Agentic AI
- Agentic AI for customer experience
- Agentic AI maturity model
- Agentic AI platform
- Agentic AI vs conversational AI
- Agentic AI vs generative AI
- Agentic Customer Experience (ACX)
- Agentic process automation
- Agentic RAG
- Agentic SOP (ASOP)
- Agentic workflow
- AI agent action
- AI agent evaluation
- AI agent for customer service
- AI agent platform
- AI agent sandbox
- AI agent vs AI assistant
- AI agent vs chatbot
- AI agent workflow
- AI agents for insurance
- AI agents vs copilots
- AI audit log
- AI concierge
- AI copilot for support
- AI coverage rate
- AI customer service agent
- AI customer service for e-commerce
- AI customer service for logistics
- AI customer service for SaaS
- AI escalation policy
- AI explainability
- AI for debt collection
- AI governance for customer service
- AI grounding
- AI guardrails
- AI hallucination
- AI hallucination detector
- AI jailbreaking
- AI observability
- AI order management
- AI quality assurance
- AI red teaming
- AI resolution rate
- AI risk scoring
- AI triage
- API support automation
- Asynchronous messaging
- Auto-Tagging
- Autonomous customer support
- Average handle time (AHT)
- Average resolution time (ART)
- Average speed of answer (ASA)
- B2B customer support
- Banking chatbot
- Bill of lading request
- Billing support automation
- BOPIS
- BPO customer service
- Bug triage
- Business logic for AI agents
- Canned response
- Carrier scorecard
- CCaaS
- Cease and desist request
- Chargeback response
- Chatbot automation
- Check call
- Churn signal detection
- Collections automation
- Communication preference request
- Confidence score
- Contact center automation
- Contact driver
- Containment rate
- Context engineering
- Context window
- Conversation analytics
- Conversational AI
- Conversational AI design
- Conversational AI in banking
- Conversational AI in healthcare
- Conversational AI in insurance
- Conversational banking
- Conversational commerce
- CSAT vs NPS vs CES
- Customer Effort Score (CES)
- Customer experience automation
- Customer health score
- Customer intent
- Customer Intent Engine
- Customer satisfaction score (CSAT)
- Customs entry status
- CX workflow automation
- Data privacy in AI customer service
- Debt validation letter
- Decision tree vs agentic procedure
- Deflection rate
- Deflection rate vs resolution rate
- Delivery exception
- Delivery order release
- Detention and demurrage dispute
- Detention vs demurrage
- Deterministic vs probabilistic agent execution
- Developer support
- Dwell time
- E-commerce customer experience (ecommerce CX)
- Ecommerce chatbot
- EDI 214
- Enterprise chatbot
- Enterprise conversational AI
- Escalation in customer support
- Escalation rate
- Exception management in freight
- FDCPA compliant communication
- Fine-tuning
- First contact resolution (FCR)
- First-party vs third-party collections
- Foundation model
- Freight claims process
- Good NPS score
- Guardrail vs guideline
- Hallucination guardrails
- Hardship support
- Helpdesk
- Helpdesk automation
- HIPAA-compliant AI
- Human agent
- Human-in-the-loop
- Input and output guardrails
- Insurance chatbot
- Intelligent virtual agent (IVA)
- Intent classification
- Intent Classification vs Keyword Routing
- Intent classifier
- Intent confidence threshold
- Intent Coverage Rate
- Intent detection
- Intent drift
- Intent Map
- Intent recognition
- Intent taxonomy
- Intent vs entity
- Intent vs topic
- Intent-based routing
- Intent-first architecture
- Intent-scoped deployment
- Knowledge base
- Knowledge base ceiling
- Knowledge management
- KYC support
- Large language model (LLM)
- Level of autonomy (autonomy spectrum) in AI
- LLM-as-a-judge
- LLMOps
- Load status update
- Loan servicing support
- Long-Tail Intents
- Model Context Protocol (MCP)
- Multi-agent system
- Multi-Intent Detection
- Multilingual conversational AI
- Named-Entity Recognition (NER)
- Natural Language Processing (NLP)
- Natural language understanding (NLU)
- Natural-language workflow authoring
- Net collections
- Net Promoter Score (NPS)
- NPS detractor
- Omnichannel contact center
- Omnichannel customer support
- Omnichannel routing
- Onboarding support
- Order cancellation automation
- Order modification
- OS&D
- Payment plan automation
- Peak season customer service
- PII redaction
- Post-purchase experience
- Pre-sale support
- Proactive support
- Product feedback loop
- Prompt injection
- Proof of delivery (POD) automation
- Provisional credit
- QA score
- Quality assurance in customer service
- Quote to book rate
- Reasoning engine
- Reg E error resolution
- Regression testing for AI agents
- Regulation F
- Reopen rate
- Resolution playbook
- Resolution rate
- Responsible AI
- Retail chatbot
- Retrieval-augmented generation (RAG)
- Returns automation
- RMA
- Self-service support
- Sentiment analysis (CX)
- Service-level agreement
- SLA compliance
- Slot Filling
- SOP for AI agents
- Speed to quote
- Staged rollout for AI agents
- Sub-intent
- Subscription support
- Support driven retention
- Support escalation to engineering
- Support queue
- System prompt
- Technical support automation
- Test-before-deploy
- Ticket routing
- Ticket volume
- Ticketing system
- Tier 1 vs tier 2 support
- Time to first response (FRT)
- Time to first value (TTFV)
- Token in AI
- TONU
- Tool calling (function calling)
- Track and trace
- Transaction dispute
- True resolution
- Trust-engineered AI
- Utterance
- Virtual assistant
- Why is deflection rate an incomplete metric
- WISMO
- WISMO automation
- WISMO in freight
- Workforce optimization
- Zero data retention