Zoom, Salesforce, Dialpad, and Others Bet Big on Agentic AI for CX

CX improvement remains the low-hanging fruit use case for all things AI, including agentic AI.

Companies compete primarily on experience today, and a single bad interaction can drive a customer to a competitor. Businesses across all industries are using AI to make the experiences of customers, students, patients, fans, and patrons memorable. Because of this, agentic AI was the top theme at this week’s Customer Contact Week (CCW) event, with vendors in the space making a flurry of announcements.

At the event, where I spent a day meeting with CX vendors and tracking the announcements, the message was clear: agentic AI is quickly becoming the contact center’s next operating layer.

Dialpad: Real-time customer intelligence with Gemini

Dialpad got a jump on CCW with the announcement last week of the integration with Google’s Gemini Enterprise for Workspace. This turns conversation intelligence, such as transcripts, sentiment, and commitments made, into a first-class data source inside Gmail, Docs, and Chat.

Users can ask Gemini to summarize recent interactions, surface risks, or prepare for a meeting using recent calls and emails, with Dialpad’s conversation data ingested natively into the Gemini environment to reduce latency and improve answer quality. This aims to address the reality that nearly 79% of opportunity data never reaches CRM systems, making the actual interaction history the source of truth for customer context.

Zoom: Agent Architect and Performance Suite

Zoom’s new Agent Architect turns prompts into production-ready voice and digital virtual agents by integrating intent, data sources, and workflows, so teams don’t have to handcraft flows. The accompanying Agent Performance Suite simulates interactions pre-launch and monitors live metrics such as containment, resolution rates, and cost per interaction, tying AI quality management to both human and virtual agents.

Zoom also introduced an enhanced customer context layer across its CX portfolio, outcome-based pricing, and multi-location deployment, enabling enterprises to build an AI service pattern once and localize it across regions.

RingCentral: Agentic AI inside RingCX

RingCentral has expanded AIR Pro to embed native agentic AI into RingCX, enabling AI agents to handle inbound and outbound interactions, autonomous outreach triggered by events, and intelligent handoffs that carry full customer history into live conversations.

A natural-language workflow builder lets operations teams design RingCX flows without code, while new analytics aim to close the gap between AI experiments and measurable customer outcomes. The company positions this as a next step in an “AI-native” strategy, backed by rising AI-driven ARR and broader investments in workforce management (WFM) and analytics.

Salesforce: WEM for Agentforce Contact Center

Salesforce, a recent entrant in the contact center space, announced Agentforce Workforce Engagement Management (WEM), which brings AI performance, human-agent metrics, and workforce management into a single view within the Service Command Center.

The platform lets leaders forecast demand, schedule both human and AI agents, and monitor adherence and quality from a single console, with native coaching tools that tie evaluations to outcomes such as CSAT and revenue. Because WEM is built directly on the Salesforce stack, it runs on the same data and workflows as CRM and digital channels, reducing integration overhead with third-party CCaaS providers.

Although Salesforce has been in the contact center space for less than a year, I expect it to close the feature gap quickly, as few companies understand customer interactions as well as the CRM leader does.

Talkdesk: Agent Builder and CXA vision

Talkdesk’s Agent Builder fits within its broader Customer Experience Automation (CXA) strategy, which proposes a unified system of autonomous AI agents that orchestrate service experiences across channels and backend systems.

CXA and its associated builder tools are positioned as a new category centered on multi-agent orchestration and low-code customization, extending agentic AI into channels like email and across business workflows. The goal is to enable enterprises to design AI agents and automation flows that span the full CX lifecycle rather than siloed bots at the edge.

8×8: AI routing and AI Studio

8×8’s latest AI routing and AI Studio releases add another dimension to agentic CX. AI Studio provides a native environment where teams can describe the agents they need in natural language and deploy them across voice and digital channels on 8×8’s existing infrastructure.

These AI agents deliver always-on inbound coverage — handling intake, identity verification, call routing, and multi-location reception — and can also drive proactive outbound engagement, sales qualification, and internal support use cases. Under the hood, 8×8’s intelligent call routing uses real-time analytics and customer profiling to match callers to the most qualified agent, improving efficiency and personalization compared with traditional rule-based routing.

What this means for AI in CX

From chatbots to agentic systems

Collectively, these announcements confirm that the contact center market is shifting from static bots to agentic AI capable of perceiving, reasoning, and acting across systems. Zoom’s Agent Architect, RingCentral’s AI agents, Talkdesk’s CXA, Dialpad’s Gemini integration, and 8×8’s AI Studio all assume AI will be a durable part of the workforce and not an add-on.

The emphasis on autonomous outreach, multi-step workflows, and multi-agent orchestration reflects a move toward large action models that can drive end-to-end outcomes rather than individual tasks.

AI as part of workforce management

Salesforce’s WEM and Zoom’s Performance Suite illustrate a second shift: AI is being managed like labor. Supervisors expect to see AI utilization, quality scores, and adherence alongside human metrics, and to forecast staffing requirements across both human and digital resources.

That changes how IT and operations approach capacity planning, service-level management, and optimization — AI becomes a schedulable, coachable resource, not just infrastructure.

Data gravity shifts toward interactions

Dialpad’s approach underscores that the richest customer signal lies in the conversation itself, not in the CRM object. By making transcripts and interaction histories queryable via Gemini and Workspace, Dialpad is shifting data gravity from static records to dynamic, unstructured interaction streams.

That aligns with Zoom’s customer context layer and RingCentral’s intelligent handoffs, both of which rely on persistent memory of past interactions to avoid “start-from-scratch” experiences.

Platform fragmentation vs unified CX fabrics

These moves also deepen the platform debate. Salesforce is betting that WEM, plus Agentforce Contact Center atop CRM, is the right nerve center for CX. Zoom, RingCentral, Talkdesk, 8×8, and Dialpad advocate their own unified CX fabrics built around contact center and communications data.

For enterprises, this raises questions about where to anchor AI governance and observability: in CRM, CCaaS, UCaaS, or a combination. The winners will likely be those who can offer clean integration patterns and shared context across these layers.

More must-read AI coverage

• SS&C Intralinks DealCentre AI vs. Datasite: Which platform is built for the future of dealmaking?
• SS&C Intralinks FundCentre AI vs. Juniper Square: Which platform better supports modern private markets fund managers?
• Why Data, Not Models, Determines AI Success
• The Rise of the AI-Native Factory: How Physical AI Is Transforming Manufacturing

Guidance for CX-focused IT professionals

Treat AI agents as products, not features

Most of these announcements emphasize faster agent buildout — prompt-based creation in Zoom and 8×8 AI Studio, and low-code flows in RingCX and Talkdesk. As an IT leader, you need product-management discipline for AI agents: clear ownership, versioning, and lifecycle management.

Define success metrics beyond containment; tie each agent to business KPIs such as NPS, first-contact resolution, and revenue impact, and use suites like Zoom’s Performance Suite, Salesforce WEM, and 8×8’s analytics to iterate.

Align AI governance with your data backbone

Dialpad’s integration with Gemini Enterprise and 8×8’s intelligent routing underscores the fact that the value of AI is constrained by data access and governance.

Before scaling agentic AI, clarify where your “source of truth” resides — CRM, ticketing, or conversation intelligence — and ensure your chosen CX platforms can ingest and act on that data under appropriate controls. This includes policies for transcript retention, PII handling, model access, and audit trails across Zoom, RingCX, Talkdesk CXA, 8×8 AI Studio, and Gemini-based workflows.

Design for a hybrid workforce by default

The future of service is a hybrid workforce of humans and AI agents. All vendors have focused on AI-to-human handoffs with shared context and common quality frameworks. Build your routing, WFM, and quality programs around that reality. That means:

• Routing policies that consider both AI and human capacity, skills, and risk thresholds.
• Quality scorecards that evaluate interactions holistically, regardless of which “worker” started or finished the engagement.
• Coaching workflows that leverage interaction data to up-level both human agents and AI policies.

Prioritize explainability and observability

Agentic AI introduces multi-step actions across systems, increasing operational risk when you can’t see what’s happening. Favor platforms that offer:

• Real-time dashboards with drill-down into AI and human performance.
• Simulation environments to test agents before they reach production.
• Clear logs of AI decisions for your risk and compliance teams to review.

Bake those requirements into RFPs and architectural standards, as “black-box” agents won’t survive enterprise scrutiny.

Build a CX AI reference architecture

Given the breadth of innovation, CX teams need a reference architecture that defines the system of record for customer data, the system of engagement for interactions, and the system of insight/action for AI.

Map which vendor plays which role, how you’ll integrate them, and how identity and context flow across the stack. Then resist the urge to “agentify” everything; identify a few high-impact journeys with clear pain points, apply these tools to design hybrid workflows, prove value, and expand from there.

Start with a few high-value journeys

Finally, resist the temptation to “agentify” everything at once.

Vendors offer accelerators, prompt-based agent creation, no-code builders, and low-friction integrations, but your organization still needs a disciplined rollout. Identify two or three customer journeys with clear pain points and measurable upside (e.g., returns, appointment scheduling, technical troubleshooting), then use these tools to design agents and hybrid workflows for those use cases first. Iterate, prove value, and expand from there.

Final thoughts

AI is no longer an add-on to customer experience platforms but rather the operating system.

The CCW announcements from Zoom, RingCentral, Salesforce, Talkdesk, 8×8, and Dialpad all point to the same destination: AI agents treated as first-class members of the workforce, governed with the same rigor as their human counterparts, and powered by deep access to interaction data. For CX-focused IT leaders, the mandate is clear.

You need an intentional architecture, a strong data and governance backbone, and a pragmatic rollout plan that starts with a few high-value journeys. Those who treat AI agents as products, not science projects, will be the ones who turn this wave of innovation into a durable competitive advantage.

You can also check out our coverage of Cisco’s warning that enterprise networks are becoming the next major AI bottleneck, as always-on AI agents place new demands on bandwidth, infrastructure, and network management.