Zerodha co-founder Nithin Kamath has warned that simply using artificial intelligence is no longer enough for a startup to stand out in front of investors. Reports from NDTV Profit and Storyboard18 frame his point as a broader shift in fundraising: AI has become common enough in startup pitches that founders must now show a stronger business, product, or execution advantage.

The remarks matter because “AI” has moved from a specialist description to a routine part of pitch decks across software, fintech, commerce, healthcare, and workplace tools. For founders, the implication is direct: presenting a product as AI-powered may explain what it does, but may no longer explain why it deserves capital, customers, or a premium valuation.

What Kamath’s warning says about startup pitches

The two reports identify the same central argument from Kamath: AI is not, by itself, a differentiator. NDTV Profit’s headline goes further by presenting AI-led pitches as potentially damaging to startups’ funding prospects, while Storyboard18 describes the issue as a failure to stand out.

That distinction is important. The reported criticism is not that startups should avoid AI, or that AI products lack commercial value. It is that the technology label has become too broad to carry a pitch on its own. An investor hearing that a company uses machine learning, generative AI, or AI agents still needs to understand the customer problem, the product’s defensibility, and the evidence that users will pay.

For a founder, that changes the order of emphasis in a pitch. AI may remain part of the product architecture, but the central argument must increasingly concern a measurable customer outcome: lower operating costs, faster work, better decisions, higher conversion, or a capability that was previously unavailable.

Evidence and limits in the available reports

The evidence for this story comes from two media reports carrying substantially similar accounts. Neither supplied in the available material a full transcript, a direct link to Kamath’s original remarks, a date for the comments, or additional context about the audience and setting. The exact wording of Kamath’s statement therefore cannot be independently assessed from the source material provided.

That limitation also applies to the stronger suggestion that AI-led pitches are hurting funding chances. The headline attributes that view to Kamath, but the available reports do not provide a funding dataset, investor survey, or deal analysis demonstrating that AI-focused companies are receiving less capital because they mention AI.

The safer conclusion is that Kamath is offering an experienced operator’s assessment of how investors may interpret increasingly similar pitches. It should not be treated as a verified market-wide benchmark. Nor does the reporting establish that Zerodha has adopted a formal rule for evaluating AI startups or that Kamath was speaking on behalf of every investor.

Still, the consistency between NDTV Profit and Storyboard18 indicates that the statement was newsworthy because it captures a concern already visible in the market: the word AI can clarify a product category while simultaneously making a company sound interchangeable with hundreds of competitors.

Why AI labels are losing their signaling power

In earlier fundraising cycles, mentioning AI could signal technical ambition or access to scarce expertise. That signal has weakened as foundation models, application programming interfaces, open-source models, and hosted inference services have become broadly available to product teams.

A company can now add summarization, search, classification, image generation, or conversational interfaces without building every underlying model itself. That lowers the barrier to launching an AI feature, but it also makes feature-level claims easier for competitors to copy. An investor therefore has to look beyond whether a product uses AI and ask what remains difficult to reproduce.

Potential answers include proprietary data, distribution, workflow integration, regulatory knowledge, customer trust, pricing discipline, or a feedback loop that improves the product with use. These are not automatic advantages, and founders still need to demonstrate them. But they are more informative than a generic claim that a product is powered by AI.

The same issue affects companies selling enterprise AI. Buyers are increasingly likely to compare tools on reliability, security, integration, governance, and total cost rather than on model terminology alone. A startup pitch that mirrors vendor language without showing operational results may therefore satisfy curiosity without creating urgency.

Implications for founders, builders, and buyers

For startup founders, Kamath’s reported position raises the bar for the “why now?” portion of a pitch. The answer cannot simply be that models have improved. Founders need to explain why the customer problem is urgent now, why their approach is better than a general-purpose model or an incumbent platform, and how the business retains value if model prices fall or capabilities improve.

Product teams face a related test. Adding an AI assistant to an existing workflow may attract attention, but adoption will depend on whether the assistant reduces friction and produces dependable results. Teams building AI agents must show where human review remains necessary, how failures are detected, and what happens when the system encounters ambiguous or sensitive tasks.

Enterprise buyers can use the same framework when assessing vendors. Instead of treating an AI feature as proof of strategic relevance, procurement and technology leaders can ask for task-level accuracy, integration requirements, data controls, auditability, and evidence from comparable deployments. The reported warning from Kamath does not prove that every AI startup is weak; it suggests that buyers and investors should demand more specific evidence.

For the wider market, the development may favor companies that combine AI with a difficult-to-replicate distribution channel or a deeply embedded workflow. It may also increase pressure on startups whose primary advantage is access to the same models and tools available to larger rivals. That does not eliminate opportunities for new companies, but it makes positioning and execution more consequential.

What to watch next

The next signal will be whether investors begin describing AI as a baseline capability rather than a standalone investment thesis. Public pitch guidance, accelerator selection criteria, and startup demo-day commentary could show whether this view is spreading beyond Kamath’s remarks.

Founders should also watch how funding conversations shift from model choice to evidence of retention, margins, deployment speed, and customer outcomes. If investors increasingly ask for workflow-level metrics, startups will need stronger instrumentation before they approach the market.

On the product side, the important test will be whether AI companies can turn technical features into durable distribution. Watch for customer references, repeat usage, measurable time or cost savings, and integrations that make a product difficult to remove. Those signals will say more about differentiation than the presence of AI in a pitch title.

Creati.ai perspective

Kamath’s reported warning is useful less as a verdict on AI startups than as a reminder about declining novelty. When a capability becomes widely available, it can remain essential to a product while losing its value as a positioning statement.

For builders and investors, the practical lesson is to separate the technology layer from the company’s actual advantage. AI may power the product, but a convincing business case still depends on a specific problem, credible execution, durable distribution, and evidence that customers care enough to pay.