Image: Wikipedia
BNN Summary
Zoho founder Sridhar Vembu has vocally questioned the prevailing AI narrative, asserting that AI is not delivering significant productivity gains and that tech companies are using it as a convenient excuse for layoffs driven by economic pressures. He also warned of an 'insane bubble' in Big Tech valuations, surpassing the 1999 dot-com era, and advocated for the shift towards private AI models to unlock their true potential.
In-Depth Analysis
Sridhar Vembu, the astute founder of Zoho Corporation, has emerged as a prominent voice challenging the often-unquestioned optimism surrounding artificial intelligence and the current valuation of major technology companies. Vembu recently stated that AI is not providing substantial productivity gains, suggesting that recent layoffs in the tech sector are more a result of underlying economic pressures and investor demands rather than genuine AI-driven automation. He posits that companies are 'blaming job losses on AI,' a move that is both 'convenient' and allows firms to appear 'visionary' in their strategic shifts.
Vembu's perspective predates the recent softening of warnings around AI's impact on employment, highlighting his early skepticism regarding the technology's immediate benefits for job displacement or efficiency improvements. He argues that the 'AI hype is like a bubble that can't fix bigger economic issues,' and that while heavy investment in AI might be temporarily buoying the US economy, this momentum is unlikely to be sustained indefinitely. He points to rising operational costs associated with building and maintaining AI infrastructure, including servers, GPUs, data centers, and advanced memory systems, as a major factor. These high expenditures, according to Vembu, compel companies to seek areas for cost reduction, with employee salaries often becoming an easy target.
Beyond AI's immediate impact, Vembu has issued a stark warning about the financial stability of the broader Big Tech sector. He characterized current Big Tech stock valuations as an 'insane bubble,' potentially even larger than the infamous 1999 dot-com bubble. To underscore his point, Vembu cited exceptionally high price-to-sales ratios for industry giants: 20x for NVIDIA, 10x for Apple, and 7.5x for Meta. He referenced a 2002 critique by former Sun Microsystems CEO Scott McNealy, who similarly admonished high market valuations of his era, drawing a parallel to the present situation. While acknowledging the 'phenomenal' growth rate of these companies, Vembu's concerns center on the sustainability of their current market capitalizations, which he believes are detached from fundamental economic realities.
Vembu also highlighted a significant shift in the future trajectory of AI development. He concurred with Oracle CTO Larry Ellison's view that AI models, such as ChatGPT and Claude, may need to transition from publicly trained systems to private, enterprise-specific models to achieve their maximum value. Vembu noted that 'AI is...rapidly commoditising,' implying that generic AI systems trained on public data offer diminishing returns. Ellison's argument centers on the idea that the real value will emerge when these models can securely process and reason over private, proprietary enterprise data, which currently largely resides in Oracle databases. Vembu added that the true 'value shifts to...what is built around it,' suggesting that specialized applications and custom solutions leveraging private data will be the next frontier. This perspective aligns with Zoho's bootstrapped philosophy and its focus on developing comprehensive, in-house solutions, including over 55 products spanning various business functions like CRM, finance, HR, productivity, and AI.
Vembu's insights extend to a broader economic outlook, where he warns of a 'grim' future and a 'difficult phase' for the global economy, urging preparedness. He views the current economic challenges as a 'gradual collapse of the post World War 2 global economic and political order,' a process he noted was already visible during the 2008 global financial crisis. He maintains that AI, despite its hype, cannot 'magically cure global imbalances' or solve deep-rooted economic problems. His company, Zoho, which he co-founded in 1996 as AdventNet, has always emphasized research and development, building its own technology stack and products completely in-house. In January 2025, Vembu transitioned from CEO to Chief Scientist at Zoho to focus more intensely on deep-tech R&D, particularly in AI development, underscoring his commitment to foundational innovation over fleeting trends.
How do you feel about this story?
