Published Oct 14, 2026 | 1:43 PM ⚊ Updated Oct 14, 2026 | 1:43 PM
Zoho founder Sridhar Vembu.
No new hiring at Zoho in the last 18 months. No pay hikes this year. And a fight to “avoid mass layoffs” at a time when the “outlook for software jobs in general is grim”.
Zoho founder Sridhar Vembu’s observations on X have kicked up a fresh debate on the future of jobs in the software industry.
Vembu posted that “as of today, AI-led productivity gains in software development estimates are (at best) 25% to 50% as measured on an end-to-end product basis (prototyping is much, much faster now but finished product requires a lot more human work).”
Going forward, though, he was confident that “a 10x or 100x productivity leap could happen”.
Vembu went on to lay out the reasons.
“Now, the 10x productivity leap in software. I believe it is coming because of what is happening in the mathematics world. The crucial link is auto-formalisation, or formal verification of code. Mathematical theorems are formalised in Lean, and proofs are verified by the Lean verifier — Lean is a programming language.
“LLMs do not verify the proof; they generate it to be verified by the Lean verifier, and once you have the advanced reasoning-capable LLM and Lean verifier in a feedback loop and give plenty of compute, magic happens…
“The analogy is handloom weaving vs powerful machine looms of code,” Vembu explained.
The Zoho founder flagged several other challenges clouding the software industry’s outlook.
Soaring server, memory, and hardware prices due to “the monster AI data centre build-out” were one concern. Then, there was the costlier AI token usage. He also mentioned how “hosting open-weight models needs expensive GPUs (read: Nvidia chips)”.
All of these factors, he noted, have driven up the costs of generating revenue growth. The state of our world is adding to the uncertainty, Vembu remarked.
His post led AltNews co-founder Mohammed Zubair to question the Zoho founder about his earlier “Come back to India” post.
“If the software industry is creating fewer jobs, and even Zoho is struggling to protect existing employment, what concrete opportunities will returning Indian professionals have?” Zubair asked.
Vembu’s response wasn’t exactly upbeat.
“When another country tells you to go home, you come home. What is the alternative? Beg them to allow us to stay? Stay illegally? There are a million things to build in India, even if software is saturated or goes away,” he replied.
How fast is the use of AI growing in the software industry?
A Goldman Sachs report captured the escalating scale with its observation that AI token consumption is set to rise 24X between 2026 and 2030, to 120 quadrillion tokens a month.
Already, this rampant dependence has been questioned by many companies, such as Uber, which saw its entire 2026 AI budget being gobbled up in four months of token use.
The Harvard Business Review quoted an AI consultant who gave the example of a client who blew away half a billion dollars in a month “after failing to limit employee licenses”. The article noted that AI token rates could rise further soon, but argued that “even at higher rates, AI usage may be very much worth the cost” as it offers multiple advantages.
In a reply to Vembu’s post too, a user, Vijay Anand, who identified himself as a former Uber and Airbnb employee, had commented on the grand AI bet most companies were undertaking. He observed that “some companies have spent so much money already (on AI) that they could have paid their employees for more than a decade with the same outcome. I am really curious to see after five years how much it was worth doing it.”
But it’s not all doom and gloom as far as predictions of the software industry go.
There are studies, like the one by the Massachusetts Institute of Technology (MIT), that found that AI automation will deliver gains in only about “23% of roles. For the remaining 77%, humans remain cheaper.”
Experts like Wharton professor Gerry Tsoukalas have also warned that fears of AI-driven layoffs might be exaggerated for another reason.
“Displaced workers are also consumers, and when their lost income is not replaced, each round of layoffs erodes the purchasing power all firms depend on. At the limit, this becomes self-destructive: firms automate their way to boundless productivity and zero demand,” Tsoukalas wrote in a recent paper with Brett Hemenway Falk from the University of Pennsylvania.
In simple terms, his argument is: What if there are no human beings to buy the products? The result is self-defeating for companies. This fear of an AI-layoff trap is too compelling, the professors believe, for a capitalist society and tech czars, who always have one eye on bottom lines and stock valuations, to ignore.
Also Read:
A Bhasmasura we built: Can Artificial Intelligence go after its makers?
Humans First, Machines Second: Reclaiming human agency in the age of artificial intelligence
Tools that enslave: Ivan Illich’s questions for India
As AI learns and grows, are we at risk of losing it all?
(Compiled and edited by R Rajesh Kumar.)
