AI is not replacing our IT engineers, making them more productive: Zoho CEO
Speaking on the edge of Zia LLM launch, Zoho CEO Mani Vambu says AI is not enough to replace engineers in his company yet.
Listen to the story
In short
- AI productivity brings benefits, but cannot replace engineers, says Mani Vambu, CEO of Zoho
- Vambu made his comment in Zoholics, where Zoho launched Zia AI
- Zia is one of the LLM based in India
In the middle of the chatter about AI replacing jobs in the technical field, there is a company that seems to be a note of optimism for software engineers. This is a zoho. Asked whether AI can replace engineers in Zoho, the company’s CEO Mani Vambu says that now the answer is no longer, as the Vambu looks at the productivity benefit from the AI tool, but he does not believe that AI can replace engineers and software developers in its company.
In a conversation with India Today Tech, Zoho CEO Mani Vambu clarified that AI is not cut on the horizon in Zoho because it is not yet enough. “If you ask (AI) system to summarize a deal or generate materials, it can do it very well. But replacing someone’s role completely? I don’t think AI is still ready for him,” he says.
And to emphasize its points, the CEO of Zoho says, “To be clear, we have not reduced any headcount due to AI. In fact, we are planning to appoint more support engineers because our support volumes are growing.”
Vambu recently made his comment in Zoholics in Bangalore. In this incident, Zoho unveiled its very large language model (LLM) called Zia. This new AI model has been picked as India’s first venture-focused, home developed AI system. Although it is not for the general public, the model is picked for businesses and is integrated with the productivity and enterprise platforms of Zoho. Available in three parameter sizes (1.3B, 2.6b, and 7b), LLM is designed to handle tasks such as structured data extraction, summary and prompt-based coding.
AI is becoming part of workflow
Vembu believes that AI can be an enabler at work, helps employees to do more. But it is not able to perform the entire AZ of tasks. “So far (in Zoho) we have definitely not seen AI in place of employees,” they say. “What we are seeing instead is this (help) in some roles. For example, a support representative usually handles 20 tickets a day, can we help them manage 25 instead of 25? This is an increase in 20 percent productivity.”
This is a point that is mentioned by AI director Ramprakash Ramamurthy in Zoho, in a special conversation with India Today Tech. He says that Zoho not only uses AI tools internally before releasing its customers, the company also knows that productivity benefits due to AI are real.
Ramamurthy says, “We dogfate all the AI projects that we are going to launch.” “Many influential areas are briefly or are searching for relevant information. For example in Zoho, a new recruitment comes in and wants to find out which form to fill for night care request. First, you have to ask someone or your boss. Now, we have HRMS products with these bots that can guide them.”
According to Zoho, a similar impact is also being seen in customer-honor roles. “Even for a seller, it is important to collect information in the system. This is where we see the most impact,” Ramamurthy says.
Asked if AI has affected the practice of keeping Zoho on work in any way, Ramamurthy does not clearly say. “For Zoho, not yet,” he says, and highlights that the company actively employs and sees AI as a tool for growth, not replacement.
There are challenges for Indian companies
While AI is helping Zoho to increase its business and make employees more productive, the company admits that Indian technical firms still face major challenges in building advanced AI systems such as Zia. Although impressive, does not look comparable to Zia Chat or Gemini’s choice in its abilities.
Ramamurthy shared that the AI space for Indian companies such as Zoho has three major challenges: calculations, data and management management.
“The first challenge is calculation. It is an expensive case, and even if you put the dollar on the table, it has a six -month lead time for it. And then there are import cap (on GPU) for countries like India,” says Ramamurthy. “The second is the availability of data. Consumers AI models work on consumer data, which is abundant. But in the enterprise, you cannot repeat it because businesses cannot only give sensitive information. The third challenge is the over-melodious of technology. It is not artificial, it is not an intelligence. Management expectations are only important.”
