Top News

Artificial Intelligence is now evolving far beyond merely answering questions, writing code, or generating images. A new category of AI—known as "AI agents"—has emerged in the tech world; these are systems designed to operate autonomously. Unlike standard AI chatbots that simply respond to prompts, AI agents can execute entire tasks from start to finish. For instance, you could ask one to find the cheapest flight, book it, and add the trip to your calendar. However, AI agents in India have faced a significant hurdle: while they can browse, offer advice, compare options, and make decisions, they hit a wall when it comes to making payments—specifically, the requirement for human authentication.

Currently, UPI payments require the user to enter an MPIN—a step designed for humans, not machines. Now, however, Pine Labs claims to have found a solution to this problem.

**Giving AI the Freedom to Complete Payments**
The company states that it has developed a system called the Pine Labs Payment Protocol (P3P), which allows AI agents to complete UPI payments without requiring repeated human approval for every transaction. This new system is built upon existing UPI payment infrastructure, specifically UPI One-Time Mandates (OTM) and Reserve Pay.

UPI One-Time Mandates were originally introduced to allow users to pre-approve future payments. This mechanism temporarily blocks a specific amount in the user's bank account for a future transaction. It is commonly used for services such as IPO applications and bookings.

Meanwhile, Reserve Pay—based on the NPCI’s "Single Block Multiple Debit" framework—was designed to facilitate recurring or automated payments. In this process, users set a spending limit upfront by entering their UPI PIN, and the exact amount is debited later based on the merchant services used. This feature is most commonly used for recurring monthly subscription payments, such as for Netflix.

Pine Labs states that these existing payment channels can now be leveraged for AI as well. According to the company, "The consumer only needs to provide authorization once at the start. Thereafter, the AI ​​agent will browse, select, negotiate, and make the payment on its own. This requires no further human approval and ensures an uninterrupted process."

How does this system keep AI in check?
This raises a straightforward question regarding security. If an AI can spend money independently, what ensures it does not exceed the set limit? Pine Labs explains that this is where 'Grantex' comes into play. Grantex essentially serves as a digital identity and permission layer for AI agents. It allows AI systems to verify their identity, grants them spending permissions, keeps them within established spending limits, and maintains a comprehensive record of every transaction.

Additionally, the system utilizes HTTP 402, which acts as a common digital language, enabling different systems to request and process payments in a standardized manner. Collectively, these technologies are designed to allow AI systems to handle purchasing and payments on behalf of a business without requiring human approval at every step.

It is already in use.
According to Pine Labs, this system is already live on the Gullak app. For instance, a Gullak user can give a direct instruction to the app—such as "Buy gold worth ₹500 if the price drops below ₹16,000 per gram"—and simply approve a UPI mandate once to facilitate this.

Disclaimer: This content has been sourced and edited from Dainik Jagran. While we have made modifications for clarity and presentation, the original content belongs to its respective authors and website. We do not claim ownership of the content.