India's Tech Pulse: Insider Daily on Startups, AI, and Big Deals

A stock trading platform founded in August 2023 just raised thirty-three million dollars at a two hundred million dollar valuation. The company is Sahi, and it’s a perfect snapshot of the speed and scale we're seeing right now. This isn't just another funding round; it's a signal. And that’s not the only signal today. The Maharashtra government just approved a ten thousand crore rupee plan to turn the state into an AI powerhouse. They're talking two thousand GPUs, six AI Excellence Centres, and a five hundred crore rupee venture fund just to kickstart AI startups. This is a state-level government acting with the speed of a venture capitalist. Meanwhile, ICICI Bank just killed the UPI PIN for small transactions—at least, for its iMobile app users.

They've rolled out biometric authentication—fingerprint or face ID—for any payment up to five thousand rupees. It’s a small change that will have a massive impact on daily user friction. And finally, in a story that feels like it’s pulled from a techno-thriller, Delhi is now actively seeking access to a super-powerful American AI model named Mythos. The catch? Anthropic, its creator, is keeping it locked down, available to only forty organizations, mostly in the US, because it’s that good at finding software vulnerabilities. Let's go back to Sahi for a second. Because the valuation isn't even the most shocking part. Between last April and this March, Sahi saw a twenty-four-fold increase in trade volumes and a nineteen-fold jump in active traders.

They just hit one million trades a day. To put that in perspective, they're already processing three percent of all daily trades in the country. How? The founders, Dale Vaz and Manish Jain, aren't just building another discount brokerage. They’re explicit that this new thirty-three million dollars is going into their AI-driven product suite. They're betting that AI can help regular traders make better decisions. They're staying laser-focused on transactional services—not getting distracted by the broader, hyper-competitive wealth management space where giants like Groww and Angel One are battling. Sahi is a bet on speed, volume, and AI-assisted execution. And so far, that bet is paying off spectacularly. But Sahi's AI focus is happening inside a much, much bigger national conversation.

And that brings us to Maharashtra and that Mythos AI model. These two stories are two sides of the exact same coin. On one side, you have the Maharashtra government’s new AI Policy. This isn't just a white paper. It's a ten thousand crore rupee—that’s over a billion dollars—investment plan. They’re creating a dedicated venture fund, offering electricity tariff concessions, and building out the actual hardware—two thousand GPUs—that startups need to train models. They’re even funding the creation of databases in Marathi. This is a government realizing that AI isn't just software; it's infrastructure, just like roads and ports. They want to create one point five lakh jobs and fifty new AI tools. It's an incredibly ambitious, ground-up effort to build a sovereign AI capability.

Then you look at the other side. While Maharashtra is laying the foundation, the central government is knocking on Anthropic's door, asking for the keys to Mythos. Why? Because Mythos is a tool that can find security holes in software before anyone else does. It's a defensive weapon, but it could easily be an offensive one. And right now, the US government effectively controls access to it. This is the strategic dilemma for India in a nutshell. You have this massive pool of engineering talent, but the most powerful, cutting-edge tools are still being built and controlled elsewhere. Waiting for homegrown models to catch up feels risky when a tool like Mythos exists today. So you have this incredible split screen. A state government pouring money into the slow, expensive, but essential work of building a domestic AI ecosystem from scratch.

At the same time, the national government is facing the raw reality of technological dependence. Sahi is building its own AI engine to win in the market. The Maharashtra government is trying to build an engine for its entire economy. And Delhi is realizing that sometimes, you can’t just buy the best engine—because it’s not for sale. It’s a race, not just for market share, but for strategic autonomy.