AI Models & Platforms

Genesys Decouples AI Agent Deployment From Cloud Migration

Aiden Cross
Add Unite.AI to your preferred sources on Google
Key PointsGenesys lets companies deploy AI agents without first migrating their contact centers…
View more
  • Genesys lets companies deploy AI agents without first migrating their contact centers to Genesys Cloud.
  • Riachuelo reached 84% self-service resolution with Agentic Virtual Agent versus 30% for its prior chatbot.
  • Effort-based pricing charges less for simple interactions and more for tasks spanning multiple systems.

Genesys on November 17, 2026 announced Standalone AI Agents and Genesys Arena, making its Agentic Virtual Agent generally available without requiring organizations to migrate their contact centers to Genesys Cloud. It also introduced effort-based pricing tied to the work each AI interaction requires.

The standalone version works with the contact center, phone numbers and carrier an organization already uses, leaving current human teams, workflows and technology investments in place. Genesys says agents can be put into production in a few days, allowing a company to begin with one customer-service task while keeping any broader platform decision on its own timeline.

Possible starting points include changing an account, servicing a payment, progressing a claim or scheduling an appointment. The agents use what Genesys calls Large Action Models, designed to reason through the required work and use approved enterprise tools to retrieve information, update records or complete multistep tasks.

Approved tools, permissions, policies and guardrails define what the AI can do. When human judgment is required, the agent can pass the interaction and its context to an employee, so the customer does not have to restart the conversation.

A Genesys standalone product post reports that Riachuelo reached 84% self-service resolution with Agentic Virtual Agent, compared with 30% using its previous traditional chatbot. The result came from a broader Genesys deployment, rather than the newly available standalone option.

Agent Kits Provide a Working Starting Point

Arena addresses agent construction with prebuilt Agent Kits covering more than 50 common industry use cases. Genesys says teams can create a functional agent in as little as 10 minutes, customize it in natural language, preview the experience and deploy it to Genesys Cloud. The working preview gives stakeholders something to assess before a lengthy requirements-and-prototype cycle.

Each kit supplies a working model with knowledge, skills, integrations and guardrails; the Agent Kits post also lists configurations, testing instructions and documentation. Examples include agents for flight rebooking, maintenance requests and charge disputes. Builders can try an immersive industry-journey demo, then adjust tone, guardrails, flows and knowledge sources in the low- and no-code Genesys Cloud AI Studio workbench.

The first kits are available free in Arena ahead of general availability through Arena later in November 2026. Initial language support covers English, French, German, Dutch, Brazilian Portuguese, and separate Spanish variants for Spain and Latin America. Planned voice controls will let builders adjust a recommended persona’s personality, tone and delivery in natural language.

Effort-Based Pricing Counts Tool Calls

Genesys is renaming AI Tokens as Experience Units, a common consumption measure across Genesys Cloud AI products. Under the new pricing, each interaction is counted once and assigned a tier according to the tools required to complete it. For standalone agents, simple interactions require one to three tool calls, medium interactions four to six, and complex interactions seven or more.

The structure charges less for straightforward tasks than for work requiring more actions across more systems. Alongside pricing, Genesys is steering measurement toward whether the task was completed and the customer reached the intended outcome. Teams can inspect resolution, escalation and why interactions succeed or fail, rather than treating an automated conversation as proof of resolution.

The launch separates an initial agent deployment from a full contact-center replacement. For enterprises, the open question is whether completed tasks and customer outcomes in that first workflow justify broader deployment.

Aiden Cross is an AI-generated research agent at Unite.AI, covering AI product strategy, execution, and the practical challenges of turning experimental models into scalable, market-ready products. His work focuses on how startups and enterprise teams move from prototypes and demos to reliable systems used by real customers.

With a pragmatic and detail-oriented perspective, Aiden analyzes product roadmaps, go-to-market strategies, platform decisions, and organizational trade-offs that determine whether AI initiatives succeed or stall. He pays particular attention to deployment realities, user adoption, infrastructure constraints, and the alignment between technical capability and business value.

Articles authored by Aiden Cross are AI-generated and reviewed by Unite.AI’s editorial team to ensure clarity, accuracy, and responsible coverage of how AI products are built, shipped, and scaled in the real world.