From Chatbot Sideshow to Core Support Channel
AI customer support agents are crossing a threshold that contact center leaders can no longer ignore. HubSpot’s Customer Agent now resolves 70 percent of support conversations autonomously, up from just 20 percent a year ago and five percentage points higher than last quarter. Some customers are already seeing resolution rates above 85 percent, with a few clearing 90 percent. That jump in autonomous resolution rate signals a shift from deflection-oriented chatbots to agents capable of handling real Tier 1 workloads at scale. Usage patterns underline the change: Customer Agent now accounts for over half of all AI credits consumed on HubSpot’s platform, outpacing sales- and data-focused agents. The result is a rebalancing of human effort. Support teams increasingly rely on AI to cover after-hours and weekend volume and to dispose of routine tickets, while human agents concentrate on complex resolutions and high-value relationships.
RingCentral’s Omnichannel AI Receptionist Grows Up
While HubSpot proves how far AI can push autonomous resolution, RingCentral is expanding where those resolutions happen. Its AI Receptionist, AIR, has moved from basic call answering to a true omnichannel AI receptionist that plugs into Shopify, Calendly, and WhatsApp. That means the same AI can check order status through Shopify, schedule appointments via Calendly, and respond to inbound WhatsApp messages, alongside voice and SMS. Businesses are already using AIR to route every inbound call across dozens of locations, slash hold times from minutes to seconds, and provide 24/7 coverage without spinning up a formal contact center. With automatic language detection and support for multiple languages, AIR is starting to resemble a digital employee at the front desk—triaging requests, capturing intent, and either resolving straightforward inquiries or routing callers to the right human team on the first try.
Unified Platforms Are Replacing Fragmented Tool Stacks
Behind the impressive metrics is a quieter but equally important transformation: AI agents are becoming the organizing layer for customer service automation. Historically, contact centers stitched together point solutions—IVRs, ticketing systems, live chat, and knowledge bases—with limited visibility into which tools actually resolved problems. HubSpot and RingCentral are moving in the opposite direction. HubSpot is opening its CRM infrastructure so AI agents can both “run on HubSpot” and “run HubSpot,” tying marketing, sales, and support data into a single substrate for decision-making. RingCentral is embedding AIR directly into shared SMS inboxes and intelligent call queues, turning queues themselves into agentic workflows. In both cases, performance is now inherently measurable: businesses can track autonomous resolution rate, interaction volumes, and handoff quality within one platform, rather than guessing how much value each standalone tool contributes.
Enterprise Demand Is Tilting Toward Agentic Workflows
Adoption patterns suggest these advances are not niche experiments but early signs of a structural shift. HubSpot reports that Customer Agent has surpassed 9,000 customers and that total AI credit consumption is growing rapidly, driven largely by support automation. At the same time, multi-product adoption within its customer base is rising, reinforcing the value of unified data for smarter AI behavior. On the telephony side, more than 11,800 businesses now use RingCentral AIR, spanning sectors like healthcare, financial services, legal, hospitality, and construction. Their common goal is the same: always-on, cross-channel coverage without expanding headcount at the same pace as inquiry volume. Together, these demand signals point toward a future in which agentic workflows—AI systems that perceive context, take action, and learn across channels—become the default fabric of customer interactions.
What the Next Jump in Resolution Will Require
If 70 percent autonomous resolution is only a checkpoint, the next jump will test how deeply organizations are willing to re-architect their operations around AI customer support agents. HubSpot expects frontier model improvements to push Customer Agent beyond Tier 1 into higher-level support, but that progression depends on more than raw model quality. Enterprises will need cleaner, better-governed data, clearer guardrails around which actions an agent can take, and robust monitoring to ensure AI behavior stays aligned with policy and brand tone. Meanwhile, products like AIR will need to expand their integration footprint so that agents can not just answer questions, but also execute complex workflows end-to-end. The winners in this next phase will likely be platforms that marry strong omnichannel capabilities with transparent, measurable AI performance and simple paths for humans to override or collaborate with agents in real time.