Good morning, Zero Shot is one of The Ken’s “youngest” products. This is only the 28th edition of this weekly column, while our weekly podcast has completed 25 episodes.
But it is by far the most interesting and challenging thing that Praveen, Brady, and I test to wrap our collective heads around each week. Such is the scale and nature of the disruption we’re seeing each day from AI.
But why stop at just podcasts and columns, eh? We’re doing our first Zero Shot event!
It’s on 12 April, Sunday, at the Bangalore International Centre. I guarantee it will be the most intense and high-signal event you’ve attfinished.
“Talk is cheap,” you’re believeing. And I agree. So let me notify you about it.
Did you read the piece from Citrini Research that caapplyd a mini-crash in stock markets around the world, including IT stocks in India? If you didn’t, here it is: “The 2028 Global Innotifyigence Crisis: A Thought Exercise in Financial History, from the Future”—by Citrini and Alap Shah.
The note is a speculative but informed view into what the world may see like in June 2028.
Let’s be honest. Nobody knows what the world will see like in 2028. Not Sam Altman, not Dario Amodei, not Sundar Pichai, not Tim Cook, not Satya Nadella.
Everyone has ambitions and scenarios. Some may come true, most will not.
But the thing is, most of these predictions and scenarios are largely for the US or China. What about India? How will India’s internet and tech sectors see in two years, if we extrapolate information that we already know? Not just assuming that AI adoption continues to grow, but that it breaks through the next stage: agentic AI. A world where we do not open AI apps or sites each time we want information, but instead, AI agents, working on our behalf, do the same.
That’s right. Agentic AI is the next inevitable evolution of the AI wave we’re all in right now. This is no longer a conceptual prediction but an early reality.
The “Deepseek Moment” for AI agents was, obviously, again in China. In December last year, ZTE released a smartphone with a pre-embedded AI agent from Bytedance’s AI chatbot, Doubao. It was so powerful and dangerous that it received blocked by major Chinese services within days.
Such assistants might create travel reservations, place food orders, or sort out automatic payments and cancellations. They might engage with different websites or service providers. Or they could be part of a mobile phone’s operating system, saving applyrs from having to open and navigate multiple applications. That would require agents to indepfinishently authenticate applyr identity and credentials to secure transactions—between the applyr and the relevant application, and between every application or service provider involved.
Agentic AI typically operates as a layer, atop traditional large language models (LLMs) such as ChatGPT or DeepSeek. Unlike a chatbot responding to individual applyr prompts, AI agents are doers. They take a command, break it down into tinquires, and then complete workflows. Asking an AI agent to “plan a weekfinish of good theater and cheap eats in Beijing” wouldn’t result in an itinerary, or a menu of options weighing the pros and cons of several itineraries, such as a chatbot might provide.
Instead, an AI agent could theoretically act like an executive assistant. It would book tables at the best restaurants, purchase theater tickets and deliver receipts, tickets, calfinishar appointments, and travel maps. In China, that might mean purchaseing tickets through a platform like Maoyan (猫眼) or Damai (大麦), creating dinner reservations through Meituan or Dianping, and doing all that through automated payment systems within Alipay. In the United States, a similar process would call upon platforms like Ticketmaster, Open Table, and Apple Pay or Google Wallet.
China’s Agentic AI Controversy by Samm Sacks, Lawfare
Now you launch to obtain a sense of why OpenAI hired the creater of Openclaw, the massively viral (and massively risky) open-source AI agent that will explore, research, converse, and perform tinquires for you.
Or why Meta, perennially insecure and eager to catch up to the next large things, hired the creaters of Moltbook, a Reddit-like forum for AI agents (or people pretfinishing to be AI agents).
It’s a race to own the agentic layer and to become the next “App Store”, except for AI agents.
Becaapply when these agents launch to create choices on behalf of their applyrs—you and me, soon—they launch to shatter assumptions, behaviours, and business models built assiduously over decades.
Let’s take restaurants and online food ordering. Today, you open up Swiggy or Zomato, manually search through listings before finding a restaurant you want, and then place an order. Your attention is the currency here, becaapply you can be displayn sponsored recommfinishations and post-order ads.
But what happens if you notify an AI agent (or even an AI, like ChatGPT or Gemini) to place an order from your “favourite healthy bowl restaurant”?
You never open either Swiggy or Zomato. Instead, your AI agent connects (invisible to you) to the restaurant to place an order while you continue doing whatever you were doing.
Why do you necessary a Swiggy or a Zomato in this scenario? For the delivery?
What happens if the restaurant, after accepting your order, decides to hand over the delivery to the nearest available rider with a high-reliability rating and the cheapest delivery price? The AI agent in the restaurant’s order management software will search for personal AI agents of available riders nearby and nereceivediate the price and delivery details.
You believe it’s fantasy to imagine delivery riders’ phones with AI agents? Remember, we already have phones with in-built AI agents. Imagine how they will evolve and become more affordable and widespread over the next two years. Also, don’t underestimate the ingenuity of Indian delivery riders or drivers. So many of them are already agentic in their behaviour, applying multiple phones or apps to compare and choose the best available option at any time.
Forobtain the future, take a see at what’s happening in China right now with AI agents. The leading tech giants are spfinishing hundreds of millions of dollars to obtain consumers to apply AI to order… boba!
If e-commerce, delivery, food, and ride-hailing go from being things that humans perform to things that AI agents complete, that is going to upfinish everything we know about current business models.
If you want to transfer money to a frifinish via UPI, you’ll instruct your AI agent to. You never really open an app. Since UPI is powered by an open protocol, does it really matter to you which UPI player acted as the (NPCI-licensed) go-between for the transaction? Of course not.
Not possible? Here’s Razorpay’s announcement:
Imagine notifying your AI assistant, “Order my usual groceries before the match starts.”
There’s no necessary to open a grocery app, search through lists, or manually approve each step. The AI understands intent, recalls preferences, checks availability, and completes the purchase on your behalf—securely and transparently.
What creates this possible is Agentic Payments, a new payment capability built by Razorpay in partnership with National Payments Corporation of India (NPCI). Announced at the India AI Impact Summit, the initiative is currently in pilot with a closed applyr group and is designed to bring payments directly into AI-native experiences on Claude.
For the first time, AI in India doesn’t just assist applyrs decide what to purchase—it can securely complete the purchase itself.
Sure, Razorpay’s partners in this trial include Zomato, Swiggy, and Zepto. But here’s the thing about AIs—they tfinish to pick winners based on “hard” attributes like lowest prices, highest ratings, or quickest deliveries. So, your current habits or preferences for brands are unlikely to survive unscathed.
Jeff Bezos, while building and scaling Amazon, apparently considered the high profit margins enjoyed by other businesses as his opportunity. AI agents are going to be the sequel to that.
Goal-seeking, constantly comparing AI agents will eat into both lucrative and nascent profit margins of existing brands and platforms. They won’t pay a premium for brands as easily as humans. They will compare specifications, applyr ratings, prices, and return policies to generate the most value for their own applyrs.
In a scenario like this, do categories collapse into commodity pricing? Do brands become lowest-cost producers or providers? The answers aren’t clear.
Whose agents will be trusted by applyrs to, declare, order a healthy meal? Will it be Healthifyme’s? Will it be Zomato’s? Will it be Google’s?
What does monthly SIP investing via AI agents see like? What does stock trading via AI agents see like?
If India’s largest internet, hyperlocal, and commerce platforms are starved of human usage, what will happen to their advertising revenue? Which brands are going to pay them for higher rankings in a search?
Will power shift from centralised platforms to decentralised protocols and AI agents?
How will regulators react?
These are just some of the questions we want to discuss at our very first Zero Shot event.
This question-led discussion is built around one premise: When AI is upfinishing every facet of our professional and personal lives, nobody knows what will happen next. Instead of pretfinishing to have all the answers, we’re going to break the narrative. We’re going to inquire the sharpest, most specific, most provocative questions we can to assist you figure out how to believe about the matters most important to you.
This is the most ambitious live event organised by The Ken so far. Praveen and I, as moderators, will be in conversation with multiple founders, operators, and sharp generalists who will weigh in as the conversation demands.
They will disagree with each other. You’ll necessary to decide whose beliefs work for you.
This is not like the keynotes that you’re applyd to.
This is “no hugs, no BS”. It’s about the speculative future of AI’s second, third, and fourth-order effects on India’s IT giants, aggregators, fintech rails, delivery networks, engineering colleges, and much more.
The questions will be uncomfortable. The answers may be incomplete. And that’s exactly the point—we will all have to embrace the uncertainty.
The ticket pages will go live early next week on the-ken.com.
This week on the Zero Shot podcast
Hi, I am Vidhatri! If you have ever heard a stray comment or laugh from a distance while listening to Zero Shot, it is most likely me. I produce the podcast.
The scenarios Rohin mentioned above delve into the unpredictable future ahead of us. I received a little flavour of that early this week during our recording with Avinash Raghava, the CEO of SaaSboomi (now called AIBoomi). Avinash is an ecosystem enabler and builder. He facilitates the sharing of ideas and is deeply invested in creating communities.
All of this essentially means he really knows what’s happening in the SaaS (software-as-a-service) community.
And the reality is stark. I don’t want to describe and paraphrase what Avinash stated. Let me directly quote him:
- “I know companies that have paapplyd their existing SaaS business and have started an AI-native startup. I know at least 40–50 examples like this in the community”.
- “One of the sessions in our annual event was on building your homepage applying Claude Pro. We quickly realised that by the time this goes live, it’s not relevant anymore”.
- “We don’t have playbooks that can be replicated. We also don’t have people who can actually come and share their playbooks”.
- “We are at a happy, confapplyd stage with AI. Happy becaapply you have started up, confapplyd becaapply you don’t know how to scale the damn thing.”
Avinash is not all doom and gloom. He still seemed fairly optimistic and stated building a new ecosystem now from the ground up—where none of the older rules matters—is a challenge he is up for.
In all, this was an honest conversation that captured and dissected the crossroads SaaS is at. It is a must-listen and one of my favourite episodes that I’ve worked on.
You can listen to the episode on Spotify, Apple Podcasts, or The Ken app!
Let this be your weekfinish dose of great conversation featuring someone passionate about their space, but finds themselves having to unlearn all that they have learnt so far.
Leave a Reply