With MCP, ACP, Google's AP2, and now UCP, the focus needs to be in laying foundations org and ecosystem wide for ease of of agentic execution across workflows, e-commerce and payments
Love this. The UCP standardization story is critical, but I think we're all missing the bigger insight buried in your conclusion:
"Friction is a feature, not a flaw."
Here's the contrarian take: The protocols that win in India won't be the most autonomous - they'll be the most *contextually* autonomous.
Fashion can get away with low-friction because wrong purchase = return it. But try that in travel (wrong flight = disaster), healthcare (wrong medicine = harm), or finance (wrong investment = broke). In these categories, intelligent friction *builds* trust, not destroys it.
So the protocol question isn't "mandate vs delegated vs real-time" - it's "which protocol allows for dynamic friction based on transaction risk?"
Example from travel:
- ₹2,000 bus ticket for frequent user → Reserve Pay, zero friction
- ₹25,000 family flight booking, first-time international → Agentic + 2FA, high friction
The magic is in the state machine: UCP gives you standardized *states*, but someone needs to build the *state transition logic* that's category-aware and context-aware.
That's the moat. Not which payment protocol you use, but whether you understand when to remove friction vs when to add it.
The fashion AI assistants proved conversational commerce works. The next wave won't be won by whoever adds AI to more categories - it'll be won by whoever understands that different categories need fundamentally different autonomy models.
Question: Have you seen anyone building this kind of dynamic friction logic? Or are we still in the "make everything frictionless" phase?
Agree. I think this will have to be built category specific first, with the 'impact' included in the context somewhere as well: as you said, a travel transaction is probably going to be more high stakes than a t-shirt.
I think right now, folks are focused more on just doing pilots / POCs of a sort to show that this works, so the 'ideal' flow is being focused on more. From a marketing / 'does the concept' work perspective, a flow without friction is way more powerful than a flow with friction to actually get people excited about this.
Not too many people (and this is my view, i could be wrong), atleast in India / SEA are focused on doing this at scale, and hence, some of these things are not being worked on. In India, with 2FA being mandatory, I think we're still focused on doing away with OTPs and introducing alternate auth like SMV / biometric creds.
Love this. The UCP standardization story is critical, but I think we're all missing the bigger insight buried in your conclusion:
"Friction is a feature, not a flaw."
Here's the contrarian take: The protocols that win in India won't be the most autonomous - they'll be the most *contextually* autonomous.
Fashion can get away with low-friction because wrong purchase = return it. But try that in travel (wrong flight = disaster), healthcare (wrong medicine = harm), or finance (wrong investment = broke). In these categories, intelligent friction *builds* trust, not destroys it.
So the protocol question isn't "mandate vs delegated vs real-time" - it's "which protocol allows for dynamic friction based on transaction risk?"
Example from travel:
- ₹2,000 bus ticket for frequent user → Reserve Pay, zero friction
- ₹25,000 family flight booking, first-time international → Agentic + 2FA, high friction
- ₹8,000 emergency same-day flight → Conversational guidance + embedded auth, medium friction
The magic is in the state machine: UCP gives you standardized *states*, but someone needs to build the *state transition logic* that's category-aware and context-aware.
That's the moat. Not which payment protocol you use, but whether you understand when to remove friction vs when to add it.
The fashion AI assistants proved conversational commerce works. The next wave won't be won by whoever adds AI to more categories - it'll be won by whoever understands that different categories need fundamentally different autonomy models.
Question: Have you seen anyone building this kind of dynamic friction logic? Or are we still in the "make everything frictionless" phase?
Agree. I think this will have to be built category specific first, with the 'impact' included in the context somewhere as well: as you said, a travel transaction is probably going to be more high stakes than a t-shirt.
I think right now, folks are focused more on just doing pilots / POCs of a sort to show that this works, so the 'ideal' flow is being focused on more. From a marketing / 'does the concept' work perspective, a flow without friction is way more powerful than a flow with friction to actually get people excited about this.
Not too many people (and this is my view, i could be wrong), atleast in India / SEA are focused on doing this at scale, and hence, some of these things are not being worked on. In India, with 2FA being mandatory, I think we're still focused on doing away with OTPs and introducing alternate auth like SMV / biometric creds.