Generative assistants

Coworker chat lets users list, filter and count orchestrated campaigns, inspect a campaign's details, lifecycle and execution history, and view its workflow as a diagram, read-only.

Coworker's Plan Campaigns skill turns a campaign brief into a full Marketo Engage program setup document with folders, smart campaigns, emails, tokens and a QA checklist.

Coworker Chat skill detects spikes and drops in ingestion, segmentation and destination metrics, reviews flow health and failed jobs; in limited availability.

Coworker Chat summarizes notifications, surfaces action items, and lets users open, mark read/unread or set reminders, with Slack delivery when connected.

CX Coworker returns read-only, profile-level execution data for published journeys: node entry counts, branch splits, errors and per-profile paths.

Adobe CX Enterprise Coworker audits the journey inventory for naming issues, stale or oversized journeys, and duplicate names, with cleanup recommendations.

Coworker chat analyzes a journey's performance and gives concrete optimization recommendations directly in the chat.

Coworker's Channel Builder skill guides administrators through creating and configuring custom channels via chat, covering authentication, throttling and payload definition.

CX Enterprise Coworker will let users run read-only SQL queries or build audiences conversationally, using Data Distiller compute hours.

Coworker will answer natural-language audit questions about Decisioning objects, including inventories, relationships and edit history.

Coworker will let marketers create and populate Decisioning offers using natural-language prompts instead of manual field entry.

Coworker will offer AI-assisted guidance when migrating from Offer Decisioning to Adobe Journey Optimizer Decisioning.

Now GA (3 Nov 2026). Current summary: Coworker chat lets users list, filter and count orchestrated campaigns, inspect a campaign's details, lifecycle and execution history, and view its workflow as a diagram, read-only.

Now GA (Nov 2026). Current summary: Coworker's Plan Campaigns skill turns a campaign brief into a full Marketo Engage program setup document with folders, smart campaigns, emails, tokens and a QA checklist.

An AI agent panel beside the journey canvas sees the open journey including unsaved edits, explains or analyzes it, and applies requested edits by highlighting changed tiles for review before saving.

Agent chat answers include inline citations after each paragraph or list item for the definitions and assumptions used, plus a full citation list at the end.

Users can add their own instructions as named custom skills and run them by name in any Breeze conversation.

Admins can describe the filter they want in plain language and Data Sync builds it automatically.

Markets are now called ICPs, built from custom keywords or found automatically by Breeze Assistant.

The Loyalty Performance view's Opportunities and Trend tabs surface AI-detected trends like tier friction or challenge drop-off with a one-click Create with AI action, and Coworker chat can generate challenge ideas.

Treasure AI Voice for iOS accepts deep links from other apps and saves their query parameters as external metadata on the resulting recording, exported to Treasure Data CDP.

Coworker Chat can produce a full-fidelity structured diff between two journey versions, including node and property changes.

Nova Agent now surfaces performance insights beyond campaign reporting and explains triggered alerts with recommended next steps.

A Coworker skill automates end-to-end journey validation (Quick Simulation flow) and helps interpret the results.

Conversational skills in CX Enterprise Coworker identify high-storage datasets, assess retention changes, and review dataset configuration and storage metrics.

A conversational skill in CX Enterprise Coworker runs statistical and semantic validations on datasets and fields to surface data quality issues via chat.

The Personalization Editor adds a 'Fix with AI' option to resolve PQL syntax errors detected during expression validation.

A Coworker skill explains, in natural language, why a specific offer was or wasn't shown to a profile or segment, tracing the decision.

A Coworker skill lets you create, explain, simulate, and optimize Decisioning eligibility rules and ranking formulas in natural language.

A Coworker skill detects anomalous spikes, drops, or flatlines in journey traffic and runs root-cause diagnostics.

Now GA (22 Sep 2026). Current summary: Conversational skills in CX Enterprise Coworker identify high-storage datasets, assess retention changes, and review dataset configuration and storage metrics.

Now GA (22 Sep 2026). Current summary: A conversational skill in CX Enterprise Coworker runs statistical and semantic validations on datasets and fields to surface data quality issues via chat.

The home page will surface AI recommendation cards when a Journey Custom Action failure or Anomaly Detected alert fires.

Now GA (7 Oct 2026). Current summary: Coworker chat analyzes a journey's performance and gives concrete optimization recommendations directly in the chat.

Now GA (20 Oct 2026). Current summary: Adobe CX Enterprise Coworker audits the journey inventory for naming issues, stale or oversized journeys, and duplicate names, with cleanup recommendations.

Journey creation with AI moves into the Coworker right rail, replacing the previous AI Assistant entry point.

The journey advanced expression editor can generate expressions from natural-language descriptions; now GA for all environments.

Now GA (30 Sep 2026). Current summary: The Loyalty Performance view's Opportunities and Trend tabs surface AI-detected trends like tier friction or challenge drop-off with a one-click Create with AI action, and Coworker chat can generate challenge ideas.

Segments can be created by describing an audience in plain language in Breeze Assistant or via Start with AI in the Segments app, reviewing suggested filters.

A new Coworker skill, Loyalty Insights, answers questions about loyalty challenge performance and program data.

Generates suggested audiences based on business context and visitor data, scoped to common retail use cases.

AI assistant in the functions code editor that writes, updates, tests and debugs functions and trigger rules from natural language.

AI assistant that generates SQL queries from natural-language prompts in the advanced query configuration for cloud data sources.

Agents can read and analyze journey definitions, runs, and logs to help debug issues or explain journey performance.

Marketers can describe a Decisioning rule in plain language in the rule builder and let AI generate it; available to a set of organizations with Adobe AI capabilities (Limited Availability).

Coworker skills let teams create loyalty challenges, set properties, and manage audiences and configuration.

An AI agent analyzes Instagram UGC for top creators, emerging themes and post performance, surfaced as proactive cards or Composer chat.

Extends the Merlin AI Campaign Insight Agent to SMS with connector performance, spike/dip, and trigger-funnel diagnostics; requires CSM access if not already enabled.

Consolidates standalone BrazeAI assistants (Liquid Agent, Copywriter, HTML Email Template agent, Image generator, Content QA, Data Transformations Copilot) into Operator.

Treasure AI Voice adds personal Summary Templates that control how call and meeting summaries, action items and titles are generated.

Merchants can ask Shopify's Sidekick assistant about Klaviyo campaigns, flows and segments and deep-link back to act.

Simulate content variations can now auto-generate realistic content variants with AI instead of building each by hand.

Now GA (1 Sep 2026). Current summary: The journey advanced expression editor can generate expressions from natural-language descriptions; now GA for all environments.

Breeze appears in global contact search results with a summary and suggested next actions, opening a pre-filled prompt when selected.

Breeze Projects can be shared with teammates with view, edit and manage permissions, with pre-configured templates and cross-session project memory.

Extends the early-access Campaign Insights Agent to Event-Triggered Email and Push campaigns with trigger-funnel summaries, pattern-finding, and diagnosis.

Six AI-powered debugging skills installable via Treasure Code generate diagnostic queries for Real-Time activation, identity and journey issues.

Now GA (26 Aug 2026). Current summary: Generates suggested audiences based on business context and visitor data, scoped to common retail use cases.

Prompt Assistant analyzes AI content generation prompts in real time, suggesting rewrites and asking clarifying questions.

AI now flags Decisioning rules and ranking formulas that can be simplified and suggests an equivalent optimized version.

Breeze Assistant asks one focused clarifying question, with options or free text, before answering instead of a wall of upfront questions.

Breeze Assistant builds charts from natural-language requests, blending CRM and external data with configurable axes, series and annotations.

Breeze surfaces summaries of companies, contacts and deals with suggested talking points ahead of meetings in the mobile calendar widget.

The HubSpot mobile app lets users attach PDFs, images and documents to Breeze conversations to summarize, extract insights, or draft replies.

A "Create report with AI" option on the Reporting page opens Breeze Assistant in a side panel to build a single-object report from a plain-language description.

Breeze Assistant answers Loop Marketing questions, walking users through defining an ideal customer, brand positioning, and campaign briefs.

Journey Agent can now generate and manage channel-specific content, such as email and push, through prompts, in Limited Availability.

Support agents now get AI-suggested reply drafts based on past tickets to review, edit and send.

Now GA (Apr 2026). Current summary: A "Create report with AI" option on the Reporting page opens Breeze Assistant in a side panel to build a single-object report from a plain-language description.

Breeze Assistant answers reporting questions that span multiple CRM objects, writing a unified query and returning a structured answer in chat.

Breeze Assistant prompts support @Mentions of CRM records, saved reusable prompts with placeholders, categorized suggestions, and a prompt library of HubSpot-created prompts and prompts shared across the team.

The Email Designer AI Assistant can use directly attached brand assets as a prompt source for generating content, instead of a separate uploaded file.

One-click AI-generated notes that summarize an audience's purpose, key criteria and intended activation.

AI-generated content can now be checked for readability, cohesiveness, and effectiveness, independent of brand guideline checks.