At Mercari India, continuous learning isn’t a box to check – it’s part of how we grow. Recently, our team had the opportunity to attend Cypher AI 2025, one of India’s largest and most forward-looking conferences on artificial intelligence. Across keynotes, panels, debates, and booth interactions, one thing was clear: AI is no longer just a technical revolution, it’s a human one.

One standout session, “Upskilling for an AI Future – Blending Tech Prowess with Human-Centric Skills,” delivered by a senior engineering leader at LinkedIn, reframed how we think about growth. The message was bold: in the AI era, degrees are good, but dynamic skills are essential. The future will reward those who stay curious, adaptable, and committed to learning. It also reminded us that while AI and data skills are in high demand, they need to be paired with empathy, creativity, and communication to create meaningful impact. As India rapidly approaches 70% AI adoption, the stakes and opportunities are higher than ever.

A thought-provoking debate titled “Generative AI and Knowledge Work – Job Enabler or Disruptor?” sparked lively discussion and sharp insights from CXO-level leaders. The key takeaway was that the shift to an AI-native organization is not about replacing people, but about transforming how work gets done, with AI as a collaborator, not a competitor.

Several technical deep dives throughout the day pushed our understanding even further. In a session titled “Co-Intelligence = IA to the Power of AI” by Tredence’s Soumendra Mohanty, we explored how to design “ambidextrous” AI systems that embrace both structure and creativity. He emphasized the need to build for ambiguity and incorporate human-in-the-loop oversight. His framework broke down GenAI integration across workflows like content creation, customer engagement, and data insights, using specialized agents to simulate real-world roles with precision.

In a session titled “The GenAI Product Trilemma,” Karan Kriplani, CPO of Neysa, outlined the constant balancing act product teams face: Unit Economics, Speed to Market, and User Trust. The trilemma suggests you can often hit two, but rarely all three. The discussion reinforced how important it is to design AI products with visibility into costs, reliable infrastructure, and robust safety frameworks from day one.

Security was another key theme. Cisco India’s Arun Shetty shared a 3-layered approach to secure AI systems throughout their lifecycle: Discovery, Detection, and Protection. From AI asset inventory and model validation to runtime guardrails, the session offered a comprehensive look at what it takes to safely scale AI across distributed environments.

In a powerful closing talk, Megha Sinha from Genpact emphasized that AI won’t scale unless we understand humans first. Her formula—Adoption = Trust × Empathy × Change—centered on the emotional intelligence and design nuance needed to truly embed AI in real-world systems. From transparency that goes beyond dashboards to cultural tailoring and social proof, the session reminded us that building trustworthy AI starts with understanding people.

Throughout the day, we also engaged with several innovative booths. From Xflow’s predictive models for currency to NxtGen’s evolving AI+infra platform, and Alstom’s AI solutions for railway systems, we saw how diverse industries are applying AI to solve grounded, real-world problems.

Our biggest takeaway? AI isn’t just an engineering conversation anymore. It’s a product conversation. A design conversation. A culture conversation. And most importantly, it’s a people conversation.

As we continue building toward an AI-native future at Mercari India, events like Cypher reaffirm why we care so deeply about learning—not just for innovation, but for transformation.