Sahi’s Dale Vaz wants an AI financial team in every pocket

Opening a trading account in India takes only minutes, but knowing what to do once you’re invested is a different matter. This gap in knowledge was the focus of Dale Vaz's keynote at TechSparks 2026. The co-founder and CEO of fintech company Sahi argued that while the last decade of fintech solved access to trading, it hasn't addressed outcomes.

Vaz observed that while most investors now have broking apps, many still lack confidence in making informed financial decisions. The "intent-capability" gap Vaz described is common among retail investors. After starting to invest, they often struggle to understand their portfolios and face anxiety when markets fluctuate. Complexity is the culprit, Vaz said, with too much jargon, data, and tools overwhelming investors.

Investors face two options: spend years becoming experts or rely on promises of guaranteed returns. While regulator-led investor education has improved, a significant gap remains between people's desire to grow their wealth and their ability to do so. AI's limitations Vaz cautioned that generic AI chatbots face three major drawbacks when it comes to financial matters.

The first is "hallucination," where a model confidently provides incorrect information. The second is "missing context," as a chatbot may respond to a question about buying HDFC stock without considering the investor's existing portfolio. The third issue is "accountability," as there is no regulatory protection or recourse if a general chatbot provides poor advice.

Sahi's solution: Kyro Sahi's answer to this challenge is Kyro, an AI assistant integrated into its app and web platform. Vaz described Kyro as a "vertical harness" around an AI model. Kyro combines a user's personal context from their activity on Sahi, 28 expert skills developed by Sahi's research team, and over 100 tools. These tools empower Kyro to perform tasks like setting price alerts, creating conditional orders, building custom scanners, and converting trading strategies described in plain language into executable code.

Crucially, the AI doesn't perform calculations; it routes requests to Sahi's backend APIs, which return accurate results. Protecting users Kyro is built upon a set of rules known as a "constitution," which mandates that answers be grounded in facts and that Kyro will always respond with "I don't know" when uncertain. Users always have the final say on buying or selling, and every output must be backed by verifiable evidence.

Vaz emphasized that AI can augment but not replace individual investment skills. Compliance and user experience Kyro complies with SEBI regulations, including in-country inference, data logging, and full traceability of conversations. Early tests with over 11,000 traders showed Kyro generating 2.7 lakh responses with a 99.2% success rate, and 78% of users returning after their first interaction.

Some users even interacted with Kyro in languages other than English, such as Gujarati, Tamil, and Marathi. What Kyro doesn't do Vaz was clear that Kyro is not an advisor, as it strictly prohibits buy or sell recommendations. To become an advisor, Kyro would need a separate SEBI investment advisory license, which Sahi plans to pursue.

Looking to the future Vaz envisions AI agents running exchange-approved algorithms for retail investors and a future where every Indian has a team of AI-enabled agents monitoring their risk, portfolio, and taxes in their native language.