Zoho is evolving its AI approach with a sharp focus on privacy and usability, ushering in what it calls the 'agentic AI era' - where automation goes beyond content creation to autonomous execution.
"Zoho's AI journey has been a steady evolution," Chandrashekar Lalapet Srinivas Prasanna (LSP), Zoho Canada's managing director, explained to TechDay during a recent interview.
"We moved from proactive to prescriptive, then generative, and now to agentic AI, which marks a significant leap in customer value."
The company first launched Zia, its proprietary AI, in 2015.
Zia was designed to enhance contextual actions across Zoho's business software suite.
By 2018, Zoho added Ask Zia - a conversational assistant capable of analysing trends and streamlining customer workflows. That laid the groundwork for deeper AI integration.
With the latest development, Zoho has rolled out Zia Agents - autonomous digital assistants that users can create and deploy with built-in skills. These agents can automate complex business processes, respond to requests, and execute tasks across departments, all without human intervention.
But while pushing forward in capability, Zoho has drawn a hard line on customer privacy.
"Our generic AI models across contextual, assistive, and agentic AI are not trained on consumer data and do not retain customer information," LSP said.
"We strike a balance between providing AI technology that assists workers while right-sizing models that don't require burdening consumers with additional costs."
This privacy-first stance comes at a time when AI use is rising, but so are concerns around data misuse. Zoho's model sidesteps this issue entirely, focusing instead on building tools with utility in mind - particularly for non-technical users.
A recent example is Zoho's "CRM for Everyone," which makes customer relationship management tools accessible to teams beyond traditional sales. Through features like Teamspaces, each department gains its own workspace inside Zoho CRM, managed independently but still overseen by a central administrator.
"Marketing teams, for example, can describe the data they need, and Zia will generate a tailored module automatically," he explained.
"This ensures CRM adoption is intuitive, efficient, and beneficial for every team - not just sales."
Ask Zia and Zia Agent Studio also allow employees to build reports, workflows, or modules without writing a single line of code. Through a chat-based interface, users can specify parameters using natural language - such as deal stages, date ranges, or customer segments - and Zia delivers actionable outputs in moments.
"Users can create workflow rules directly within the Ask Zia chat window," said LSP. "They can configure actions like assigning owners, triggering email notifications, and invoking custom functions without writing code."
The aim is to give all businesses - regardless of size or budget - the ability to work smarter with AI. According to LSP, that means embedding AI features directly into existing software rather than charging extra for them.
"We integrate AI capabilities directly into our products without requiring separate subscriptions or costly add-ons," he said.
"This ensures businesses of all sizes can leverage AI to understand workflows, anticipate needs, and automate tasks."
Zoho has also launched Connected Records and Connected Workflows to boost collaboration across departments. These features allow data to flow seamlessly between teams, automating follow-ups and transitions from sales to support.
"Connected Records allow teams to link related data across CRM modules," he said.
"Sales, marketing, and support teams can gain contextual insights without switching systems."
Meanwhile, the company's CoCreator feature in Zoho Creator brings app-building capabilities to non-technical users. Through natural language, diagrams, or specification documents, users can describe their vision and Zia will generate a functional application with dashboards, modules, and integrations.
"Even users with little or no coding expertise can generate tailored code blocks automatically," LSP said.
"For those refining existing applications, CoCreator optimises and annotates code for efficiency."
For partners and developers, Zoho offers an Agent Marketplace, which allows system integrators and resellers to sell or deploy AI agents tailored to specific industries. These agents can be customised without code, speeding up client implementations and enabling partners to offer packaged AI solutions.
"Partners can develop and sell custom AI agents," he said. "It empowers integrators and resellers to scale their businesses, enhance customer experiences, and drive AI adoption across industries."
In the Canadian market, however, challenges persist.
According to Zoho's Digital Workplace Transformation study, many companies lag behind global peers in AI adoption and security practices.
"Eighty-five percent of Canadian companies still rely on manual task delegation rather than automation," LSP said.
"Fewer than half enforce strong authentication like multi-factor or biometric access."
Zoho sees this as an opportunity. By combining AI with automation, integration, and privacy-first security, it aims to help Canadian businesses close the gap.
LSP concluded by highlighting the company's full-stack control - from infrastructure to applications - as key to delivering value.
"Zoho's distinct technological breadth and depth is a fundamental differentiator," he said.
"Our highly secure, privacy-compliant, capable, and deeply knowledgeable AI agents deliver superior technology at a high value to our growing customer base."