The conversation was sparked by recent developments in generative AI and particularly the rise of models like DeepSeek, prompting Patil to reflect on whether such tools pose a threat to existing LLMs
In a candid conversation that spanned across artificial intelligence, cybersecurity, and blockchain, Dr. Satish Patil, the founder of Kanlet Technologies and a serial entrepreneur, offered sharp insights into the evolving tech landscape. Known for his technical depth and practical understanding, Patil spoke on the transformative impact of Large Language Models (LLMs), the myth surrounding zero trust architecture, and the possibilities that quantum computing could unlock in the blockchain space.
The conversation was sparked by recent developments in generative AI and particularly the rise of models like DeepSeek, prompting Patil to reflect on whether such tools pose a threat to existing LLMs.
“I definitely do not see a threat,” he said, adding, “LLMs have entirely changed the game in terms of how we access and interact with information. The key challenge is not the existence of new models, but how these models are trained and used—especially in regulated sectors like healthcare and finance.”
He noted that while errors in areas such as entertainment or retail recommendations are tolerable, the same cannot be said for fields where accuracy and reliability are non-negotiable. “You cannot go wrong with disease identification or treatment recommendations,” he emphasised.
Beyond Silos: Future Of Agent-To-Agent Communication
Patil offered a vision for the future of LLMs that goes beyond current limitations. He believes the next leap will come when LLMs, or AI agents, start interacting across ecosystems. “Today, an agent from Gemini does not talk to Microsoft Copilot. But tomorrow, you’ll see bots from Salesforce, Google, Microsoft and even X (Grok) interacting seamlessly,” he predicted.
This agent-to-agent interaction, he argued, is not just about improving intelligence or productivity. “It’s about enabling real enterprise use. Big companies run on diverse tech stacks. If a powerful agent exists on Google Cloud, but the company is on Microsoft Azure, they simply can’t use it. We need cross-platform agents to fully utilise enterprise data and serve cross-functional use cases.”
According to him, the future lies in creating a marketplace for such agents—akin to SaaS today—where users can select agents from multiple platforms to build their own AI workflows. “Today, these agents exist in silos. But in the future, it will be about assembling an ecosystem with different agents talking to each other,” he said.
On Zero Trust: “It’s Myth, Not Milestone”
Shifting focus to cybersecurity, Patil offered a refreshing take on the much-discussed zero trust infrastructure , a concept that has been gaining traction over the past decade but remains nebulous in its implementation.
“You’re absolutely right—it was relevant in 2015, in 2020, and it’s still relevant in 2025. Believe me, it will be relevant in 2035 too,” he said, laughing.
For Patil, zero trust is more of a moving target than a tangible achievement. “Every day, new data is generated, new use cases appear. Security frameworks must evolve in real-time. That’s why I call it a near-zero trust infrastructure, not zero trust. There’s always a delta of unknown information.”
When asked if he meant compliance-based strategies or broader innovation, he clarified, “I’m not talking about compliance. I’m talking about actual implementation. We should think in terms of close-to-zero trust, much like near real-time in computing—absolute zero trust is a myth.”
Quantum’s Role In Blockchain: Risk Or Catalyst?
Patil concluded with his views on blockchain—a domain often caught between hype and hesitation. While many experts remain sceptical about how quantum computing may disrupt blockchain’s cryptographic foundations, Kumar took a more nuanced view.
“Some say quantum is a threat to blockchain, but I believe it will actually help. One of the biggest challenges in blockchain today is setting up the tech infrastructure. Quantum will reduce the barrier to entry and allow for the creation of micro, vertical-focused blockchains,” he explained.
At present, blockchain adoption is largely horizontal and generic. But with quantum capabilities, he predicts a future where niche blockchains—focused on specific sectors or use cases—can emerge more freely. “Usage is commoditised, but creation is not. With quantum, more people will be able to create blockchains for specific needs.”
What made this interaction particularly engaging was patil’s clarity and command over the subjects. Rarely do you find startup founders with such technical depth, especially those who’ve built multiple successful ventures. As the conversation wrapped up, his pragmatic optimism resonated well—highlighting that while technology evolves rapidly, it must be built on foundations that are both thoughtful and adaptive.
As Patil succinctly put it, “People have to implement. People have to learn. But there will never be a situation where we say, ‘Yes, I’ve implemented zero trust.’ It’s always evolving.”
And perhaps that’s the best way to look at emerging technologies—not as destinations, but as directions.
