Zoho CEO: 'Vibe Coding' Will Fail

Vembu's stance is rooted in Zoho's long-standing, unconventional approach. Founded in 1996, the company has been bootstrapped since day one, deliberately avoiding venture capital to maintain control and focus on long-term R&D. This philosophy of patient, deliberate building informs his skepticism of AI hype cycles. Developer sentiment appears to be aligning with Vembu's caution. A recent Stack Overflow survey showed a significant drop in how favorably developers view AI tools in their workflow, falling from 72% in 2024 to 60%. Trust in the accuracy of AI-generated code has also declined, with only 33% of developers expressing confidence, down from 43% the previous year. While 72% of IT professionals report using AI coding assistants daily, a staggering 96% do not fully trust the output to be functionally correct. A primary concern is that AI-generated code often looks correct but is unreliable, leading to a new bottleneck: developers spend more time reviewing AI output than it would have taken to write the code themselves. This skepticism is fueled by tangible issues. One study found that 57% of engineering teams saw a net negative effect on productivity due to the extra work needed to maintain AI-generated code quality. Security is another major concern, with one analysis finding that nearly half of all code generated by certain AI tools contained known security vulnerabilities. Vembu himself is not entirely against AI, but advocates for a more measured application. Within Zoho, engineers are permitted to use AI tools but must review and take full responsibility for all generated code. He sees AI's immediate value in accelerating specific tasks, like user interface development, where one experienced engineer reported completing three weeks of work in a single day by using AI to flesh out a human-provided structure. He argues that true engineering skill lies in "context-free grokkability"—the ability to understand a random snippet of code knowing only the domain. Vembu believes AI may be a double-edged sword, boosting the productivity of senior architects who provide the necessary context and judgment, while diminishing opportunities for junior engineers who learn by working through that "boiler plate" code. Ultimately, Vembu champions a vision of AI that is more sustainable and efficient, criticizing the massive energy footprint of current large-scale models. His focus remains on deep R&D and technological sovereignty, arguing that India needs more patient, bootstrapped companies that control their own core technologies. This long-term view prioritizes durable engineering over chasing short-term, AI-driven productivity gains.