Delight.ai vs Sinch (2026)

Delight.ai integrates with Zendesk, Salesforce and Freshworks for live agent handoff, with Salesforce also supported as a knowledge source and third-party knowledge base integration so the agent can surface the right information in real time. Agents take action as well as answer, and Trust OS logs every tool call. On the channel side the agent reaches customers through in-app chat, web, email, SMS, WhatsApp and social, all built on Sendbird's communications platform, and an AI Agent Integrations page carries the connector set.

SourceDelight.ai integrations page, AI agent platform page and Salesforce and Freshworks product releasesread 2026-09-05

Action blocks connect a bot to outside systems in three ways. An API step sends requests to any HTTPS endpoint with any of the five standard HTTP methods, authenticates with basic auth or a bearer token, passes session variables in the query or body, and can route the conversation to the next block named in the response. Code actions run JavaScript that calls external services and writes the results back to the session.

Native app steps create, find and update Zendesk tickets and users, create Salesforce contacts, accounts and cases or run SOQL queries, and connect Freshdesk, Topdesk, Gmail and Intercom. In a banking bot, an API step can post a money transfer's amount, destination and account type to the customer's back end in the middle of a conversation.

Sourcedocs.chatlayer.ai/integrateandcode/custom-back-end-integrations/configure-your-api-integrationread 2026-10-11

Agent Steward takes full ownership of complex, long-running customer issues end to end instead of answering and closing, and its agents understand intent, hold context, take action and adapt, resolving complex issues and escalating only when needed.

Behavior comes from a customer-authored actionbook holding rules such as refund thresholds and escalation logic, which Openbook renders as a visual flow through complex branching.

Above the individual case, Zero-Touch Improvement runs a closed loop in which observability catches an issue, control tests the fix and oversight holds it to policy, in a graduated system that moves from human-guided to autonomous.

SourceDelight.ai Agent Steward, Trust OS and Omnipresent pagesread 2026-09-05

Bots run as flows on a visual canvas, where each block sends a message, collects input, checks a condition or runs an action, and go to connections move the conversation between blocks and flows.

Condition blocks branch on variables, an API response can name the next block, events start a flow when a variable changes or the user goes quiet, and delay and pause steps hold the conversation. Flows can be reused across a bot and published one at a time. The intent engine picks the flow and the designer sets every step inside it, so the sequence is fixed when the bot is built. Running two bots in one conversation takes the web widget or a custom proxy over the webhook channel.

Sourcedocs.chatlayer.ai/navigation/bot-builder/flowsread 2026-10-11

Channels are in-app chat, website, email, SMS, WhatsApp, social messaging and voice, all carried on Sendbird's communications platform, with context following customers between channels instead of restarting. Omnipresent agents are always on, always aware and always personal, and they start conversations themselves, anticipating needs before an issue forms and re-engaging when a customer goes quiet or switches channel. Voice 2.0 adds multilingual real-time outbound calling.

SourceDelight.ai Omnipresent, Voice AI Agent and AI agent platform pagesread 2026-09-05

Conversations arrive through the Sinch Conversation API on WhatsApp, RCS, SMS, MMS, Messenger, Instagram, Viber, Telegram and LINE, through the Web V2 widget, through a webhook channel for any custom front end, and through phone lines for voicebots.

Inside a conversation, events start a flow when a variable changes, when the user stays silent for anywhere from 1 to 1,440 minutes, or when a WhatsApp Flow is completed.

Outside systems can also push into a live conversation through the REST API, sending messages, setting session data or triggering a specific block for a given user, so a back end event can move a conversation forward.

Sourcedocs.chatlayer.ai/channels/all-channelsread 2026-10-11

Delight.ai integrates third-party knowledge bases so the agent can find and surface the correct information in real time, with Salesforce supported as a knowledge source and Freshworks and Zendesk connected alongside. Building an agent works like onboarding a team member, transferring knowledge to it through the supervisor dashboard, and the actionbook holds the authored rules the agent answers within. Trust OS detects factual errors and off-policy responses automatically, and the review-and-grade surface scores knowledge accuracy.

SourceDelight.ai AI agent platform and builder pages, and the Salesforce knowledge-source releaseread 2026-09-05

Knowledge base AI keeps a store of the customer's content and generates short answers from it. Content goes in as PDF files, CSV question and answer files, single web pages or entire domains, and tags limit which content a given flow searches.

Each answer comes back with the retrieved passages, the source file or URL and the specific page it came from, so a bot can offer a link to the source. When nothing matches, a no findings path sends the user to another block instead of inventing an answer, and a non generative mode replies only with the exact answers in a CSV file. The feature is free up to 50 questions a month, with higher volumes through sales.

Sourcedocs.chatlayer.ai/navigation/knowledge-base-airead 2026-10-11

The Agent Memory Platform, sold as its own product, is the intelligence layer that gives agents a living memory and a business brain, connecting what the agent remembers about each customer with what the business is trying to achieve.

Agents absorb signals from every conversation and action to form an evolving understanding of the customer that goes beyond static CRM data, and context is kept as a customer moves between chat, SMS, email and voice so nobody repeats themselves.

The agent remembers every interaction, and continuity carries across separate conversations as well as within one, unlike conventional chatbots that lose history the moment a session ends.

SourceDelight.ai Agent Memory Platform and AI agent platform pagesread 2026-09-05

Each user's session holds what the bot has learned about them, including variables it collected, API results and the last intent, and one session can span several conversations. Sessions expire after 30 days without activity by default, a period Enterprise customers can change per bot and separately for the draft and live versions.

Through the REST API a team can read a session, write to it, delete it for one conversation or delete all data for one user. Bots use saved variables to recognize returning users and reuse earlier answers, such as a traveler's usual route. Persistence depends on the channel, and the web widget keeps variables for one visit unless users are identified. Variables marked sensitive stay out of conversation history.

Sourcedocs.chatlayer.ai/bot-answers/sessionread 2026-10-11

Trust OS has kill switch and escalation controls that pause, redirect or kill any AI flow instantly, with one-click escalation to a human agent and role-based access control so the right people own the right agents. Policy is customer-authored through the Actionbook Editor, which lets operations teams edit refund thresholds, escalation logic and tone in real time with no code and no engineering ticket.

Hallucination and unsafe content detection runs automatically, flagging factual errors, off-policy responses and safety violations before customers see them, and a review-and-grade surface lets people score responses and flag outliers against the customer's own standards. Handoff to a person runs through the Zendesk, Salesforce and Freshworks integrations. Oversight comes through authored policy, automatic detection and interruption, not a sign off on each action before it is sent.

SourceDelight.ai Trust OS and AI agent platform pagesread 2026-09-05

A Send to offload provider step hands a conversation to a person in Sinch Contact Pro, Genesys Cloud, Salesforce Service Cloud, Zendesk Chat, Help Scout or any platform behind an offloading webhook, with the transcript and collected variables passed along. The bot keeps answering until an agent accepts, then pauses until the agent sends the user back.

Designers choose where handover happens, typically on request, after repeated not understood turns or after two unsatisfactory knowledge base answers. Generative features can be switched off per bot, and a split between draft and live versions keeps unpublished changes away from customers. No step lets a person approve an action or a reply before the bot sends it.

Sourcedocs.chatlayer.ai/integrateandcode/human-offloading-live-chatread 2026-10-11

SOC 2 Type 2, HIPAA, GDPR and CCPA are named on the security page, and ISO 27001 and HIPAA/HITECH on the platform page, with a live trust center at trust.sendbird.com.

Controls the customer sets include granular role-based access controls, full audit logs, flexible data retention settings aligned to its own policy, end-to-end encryption and TLS in transit, PII protection and data masking, and RBAC also scopes which people own which agents.

Internally, Sendbird uses FIDO2 multi-factor authentication with device trust, automated detection and response, automated compliance and a DevSecOps development approach, and it publishes a vulnerability disclosure address.

SourceDelight.ai security page, AI agent platform page and Trust OS pageread 2026-09-05

Sign in runs through Sinch ID, and SAML 2.0 single sign on with Azure Active Directory, Okta, OneLogin or Ping Identity comes with the Enterprise pack. Admins see every bot and manage the team, members see open bots plus the restricted bots they are granted, and each restricted bot gives a member one of five permissions, Edit, Read only, NLP only, Bot Builder or Copywriter.

Legacy password accounts lock after 10 failed sign ins. Chatlayer is ISO 27001 certified, and across its products Sinch holds ISO 27001:2022 and a SOC 2 Type II report, with PCI and HIPAA compliance covering specific products. Security reports and 343 answered questionnaire items sit in the Sinch Trust Center behind an access request.

Sourcedocs.chatlayer.ai/support/access-controlread 2026-10-11

Trust OS logs everything for tracing, and Openbook logs every conversation, every decision and every tool call for review, with no black boxes. Openbook also renders the agent's actionbook as a visual flow, so an operator can follow the logic behind a decision. Low CSAT Analysis sorts every conversation scored one or two into one of five failure types before anyone opens it, and Thinking Messages show a status for each tool call during the run, so the reasoning is visible as it happens as well as afterward. A dedicated Analytics product tracks conversations, common topics and ROI.

SourceDelight.ai Trust OS and Analytics pagesread 2026-09-05

Conversation history keeps a turn by turn transcript of every conversation with the intents triggered along the way, voice recordings where recording is on, and tags and filters by user, channel, campaign and date.

Execution logs record each app integration run with its input and output payloads, status, account and the block that triggered it, plus failed API and code actions with the request, status code and response, tracking events and voice speech to text failures. An activity log on the Enterprise pack shows who accessed or changed a bot, what they did and when, and every published live version is kept for rollback. A conversation flow view, labeled beta, traces a conversation's path through blocks for debugging.

Sourcedocs.chatlayer.ai/navigation/history/execution-logsread 2026-10-11

Customers keep full control over where their data resides, how it is stored and who can access it, to meet regulatory and internal data governance requirements, with regional hosting to support data residency needs. The platform lists region-specific storage among its privacy-first data governance properties, alongside PII protection and data masking. Retention is customer-set, with flexible data retention settings aligned to the customer's own policy, and data ownership and processing are described as transparent. No on-premise or self-hosted deployment is offered.

SourceDelight.ai security and AI agent platform pagesread 2026-09-05

Bots run in one of three regions on Google Cloud, each with its own app URL and sign in endpoint, in St. Ghislain, Belgium for the EU, Ashburn, Virginia for the US and Mumbai, India for Asia. Moving bots between regions goes through support, and fixed ingress and egress IP ranges per region let customers allowlist Chatlayer in their own networks.

Sinch's subprocessor register adds Brazil and India as Chatlayer hosting locations, and the Tables API has a South America endpoint, so some data can sit outside the three bot regions. Generative replies through the OpenAI integration send conversation data to OpenAI services in the United States.

Sourcedocs.chatlayer.ai/support/saas-regions-and-ip-rangesread 2026-10-11

Capabilities each have their own page and are adopted as units, among them Agent Steward for autonomous ownership of long-running issues, Agent Memory Platform, For You Conversations, Omnipresent, Voice AI Agent, Trust OS, Analytics, Desk and AI Agent Integrations. Industry pages cover retail, travel, healthcare, financial services, logistics and more.

There is no template gallery, agent library or marketplace for customers to browse and put into service, and agents are built by onboarding them like a new team member, transferring knowledge and providing workflows through a supervisor dashboard, with behavior authored in an actionbook.

SourceDelight.ai AI agent builder, capabilities index and industry pagesread 2026-09-05

New bots can start from templates built by Chatlayer's conversation design team, each a working bot for one job.

The set covers a banking assistant that makes payments, checks balances and card limits and reports lost cards, ecommerce returns, an electric bike shop that books test rides and checkups, shared scooter support, a bilingual gym assistant that books and cancels classes, job applications, lead generation, NPS surveys, restaurant and retail bots, a nearest location finder and a knowledge base AI bot.

Most work as soon as they load, while those using an integration need an account connected first. Smaller templates add a single function, such as feedback collection, to a bot that already exists.

Sourcedocs.chatlayer.ai/start-quickly/bot-templatesread 2026-10-11

Delight.ai does not say which models run the agent or who chooses them, and there is no model selection, routing configuration or bring-your-own-key path. The product is sold as a managed outcome, with resolution rates, a governance layer that detects hallucinations, and zero-touch improvement that tunes the agent on its own.

SourceDelight.ai AI agent platform, Trust OS and security pagesread 2026-09-05

An OpenAI action lets a builder pick GPT-3, ChatGPT or GPT-4 for a generated reply, using Chatlayer's own key for the first 100 messages per bot and the customer's OpenAI key after that.

The other generative features, knowledge base AI, the AI intent booster and expression generation, run on unnamed models Chatlayer picks, with OpenAI and Microsoft Azure as Chatlayer's machine learning subprocessors. Intent recognition runs on Chatlayer's own language engine, trained on each customer's expressions.

OpenAI is the only model provider in the integration catalog, and the model list in the OpenAI action still reflects the GPT-3 era.

Sourcedocs.chatlayer.ai/integrateandcode/app-integrations/openai-gpt-3-chatgpt-and-gpt-4read 2026-10-11

Sendbird publishes AI agent documentation at sendbird.com/docs/ai-agent, linked from delight.ai's developer menu alongside a public community forum and a server status page, plus a GitHub organization at github.com/sendbird.

The underlying communications platform ships cross-platform SDKs for iOS, Android, JavaScript, React Native and Unity with chat, voice and video APIs, which embed agents natively in a customer's own app instead of as a widget.

AI Agent Integrations connect the agent outward to Zendesk, Salesforce and Freshworks, and Trust OS exposes the actionbook as an editable artifact that operations teams author directly. There is no MCP server or agent manifest.

SourceSendbird AI agent documentation, delight.ai developer menu and the Sendbird communications API platform; sendbird.com/docs/ai-agent/guide/v1/overviewread 2026-09-05

A public REST API with an OpenAPI 3.0 specification covers messaging, conversations, bots and NLP.

It sends messages through the webhook channel with asynchronous callbacks, lists conversations and messages, reads and deletes session data, deletes a user's data, executes actions in a live conversation, lists bots and channels, creates webhook channels and returns the NLP result for any expression. Calls authenticate with bearer tokens generated under Settings.

A separate Tables API, with basic authentication and regional base URLs, loads and queries the built in tables that bots read and write, and whole bots export and import as files.

Sourceapi.chatlayer.ai/v1/docsread 2026-10-11

Every change is tested before it touches a customer, with staging, real-conversation test suites, gradual rollout and A/B data, so a deploy comes with evidence. Staging environments mirror production with real integrations and policies. Test Upgrade builds test cases from real conversation IDs, organizes them with labels, shares sets across environments and returns consistent pass/fail results.

Gradual rollout ships to ten percent first, and A/B testing runs two versions head to head for hard data on resolution rate and CSAT before committing. Simulated environments test and evaluate an agent before deployment, and the Trust OS review-and-grade surface gives every response a quality score.

SourceDelight.ai Trust OS and AI agent platform pagesread 2026-09-05

An NLP dashboard rates each language's intent model from 0 to 100 percent, shows how that rating moved since the last training, and flags intents with too few expressions, misclassified expressions and overlaps between intents.

After launch, the Train tab collects real customer messages with the suggested intent and a confidence score, and the team corrects them and adds them to the model ahead of the next release. An emulator with a debugger shows session data turn by turn, voicebots and a WhatsApp sandbox can be tried before going live, and a bad release rolls back to an earlier live version. Saved test cases that replay flows, actions or knowledge base answers are not part of the product.

Sourcedocs.chatlayer.ai/navigation/natural-language-processing-nlp/dashboardread 2026-10-11

The agent acts through integrations with Zendesk, Salesforce and Freshworks, through tool calls that Trust OS logs individually, and through the communications APIs and SDKs that embed it in the customer's own app. It talks to people over in-app chat, web, email, SMS, WhatsApp, social and voice, and Delight.ai has no browser control, hosted session, virtual desktop or interface automation.

SourceDelight.ai AI agent platform and Trust OS pagesread 2026-09-05

Bots work through API calls, JavaScript code actions and native app integrations, and voicebots handle phone calls with speech recognition and synthesis. Knowledge base AI crawls a page or a whole domain when content is added, which builds the answer store before any conversation takes place. Chatlayer has no hosted browser, remote desktop or screen control, and Sinch's tools for agents built elsewhere work the same way, since the MCP server, Functions and Voice Relay all act through Sinch's own APIs.

Sourcedocs.chatlayer.ai/buildabot/flow-logic/dialog-state/action-bot-dialogread 2026-10-11