Pine Labs Launches Agentic Payments Protocol P3P

Pine Labs announced the Pine Labs Payment Protocol (P3P) on June 11, 2026, a system that lets AI agents complete UPI payments after a single upfront user authorisation, according to the company's press release and multiple media reports. P3P combines UPI mandate frameworks (Single Block Multiple Debit / ReservePay and One Time Mandate), an identity and delegation layer called Grantex, and an HTTP 402 machine-readable payment request standard to let third-party agents pay within pre-approved limits. Pine Labs and several outlets report Gullak is live on P3P and Vijay Sales is running a proof of concept. Reporting by Medianama and IndiaToday highlights open questions about liability, consumer privacy, and how existing RBI/UPI rules apply. Editorial analysis: Industry observers will be watching regulatory guidance, merchant settlement flows, and how audit logs and revocation controls perform in production.
What happened
Pine Labs announced the Pine Labs Payment Protocol (P3P) on June 11, 2026, describing it as a protocol that allows AI agents to complete UPI payments after a single, upfront user authorisation (Medianama; Business Standard; ANI). Pine Labs frames P3P as extending existing UPI mandate rails, Single Block Multiple Debit (SBMD), marketed by Pine Labs as UPI ReservePay, and One Time Mandate (OTM), so an agent can trigger debits against funds that were reserved or pre-authorised (Medianama; Pine Labs press release). Pine Labs and multiple news outlets report Gullak is live on P3P; Vijay Sales is testing a proof of concept (Business Standard; ANI; EconomicTimes).
Technical details
Per Pine Labs' documentation and press materials, P3P combines three components:
- •UPI mandate frameworks (SBMD/ReservePay and OTM) to set aside funds or pre-authorise a spend;
- •Grantex, described by Pine Labs as an identity/delegation layer that verifies agent identity, enforces spending limits, and records transactions;
- •HTTP 402, an open machine-readable payment request standard enabling agents on different platforms to discover and request payment uniformly (Pine Labs press release; Medianama). Pine Labs states: "The consumer authorises once, upfront. After that, the agent browses, selects, negotiates, and pays. No human authentication. No interruption. No friction." The company also lists cards, net banking, wallets, EMI and stablecoins as future payment rails (Pine Labs press release; Medianama).
Early adopters and use cases
Gullak co-founder Manthan Shah confirmed the integration to ANI: "With P3P, we are moving from manual savings to autonomous wealth creation. Pine Labs has built the infrastructure to make this possible." Other use cases include tracking prices of flights, precious metals, or concert tickets and autonomously executing a transaction when target conditions are met (ANI; Business Standard). Pine Labs states deployments are underway across retail, fintech, and travel (Pine Labs press release).
Scale anchors
Pine Labs' press release cites UPI volumes, noting UPI processes over 2.3 billion transactions a month (NPCI, May 2026) to frame the opportunity for agentic payments in Indian commerce (Pine Labs press release). Multiple outlets emphasise that P3P extends existing mandate constructs rather than replacing UPI authentication flows (IndiaToday; EconomicTimes; Medianama).
CEO positioning
CEO Amrish Rau framed the launch as India-led standards-setting, telling Business Standard: "India shouldn't be looking to the West to figure out what is the protocol they are doing and how this can be implemented." Rau confirmed the firm is in discussions with card networks to extend P3P beyond UPI, and stated the protocol is "completely compliant" with existing regulatory guidelines (Business Standard).
Industry context
Editorial analysis: Companies building agentic commerce systems commonly adopt mandate-style pre-authorisations to enable autonomous actions while preserving user control, since continuous human authentication is impractical for machine-driven workflows. Combining a delegation/identity layer with auditable logs is a standard approach to balance autonomy with traceability in agentic systems.
Regulatory, liability and privacy questions
Media reporting raises open questions. Medianama frames the launch as prompting questions about how UPI rules, liability allocation, and consumer privacy apply to agent-initiated debits (Medianama). IndiaToday and other outlets note that RBI and NPCI guidance specific to autonomous agent payments is not yet defined, leaving open questions around dispute resolution and merchant/agent liability (IndiaToday; EconomicTimes). Reporters also flag data-retention concerns around the Grantex audit trail (Medianama).
What to watch
Editorial analysis: Practitioners and observers should track:
- •whether NPCI or RBI issue clarifications addressing agent-initiated mandate use-cases
- •how banks and PSPs implement SBMD/OTM flows for agent-triggered debits and whether settlement/liability flows differ
- •whether independent audits or incident disclosures reveal gaps in Grantex authentication or revocation controls. Engineers building agentic workflows should validate revocation latency, audit-log tamper-resistance, and UX around mandate expiry and cancellation
Bottom line
Pine Labs has introduced a production-grade protocol that uses existing UPI mandate primitives plus an identity layer and HTTP 402 to let AI agents pay within pre-approved boundaries. Reporting from Medianama, Business Standard, EconomicTimes, and ANI confirms real deployments are live, while raising unresolved regulatory, liability, and privacy questions practitioners will need to monitor.
Scoring Rationale
A notable production launch enabling agentic commerce on India's UPI infrastructure, with a confirmed live deployment (Gullak) and credible CEO positioning around India-led standards. Directly relevant to engineers building agentic workflows and fintech practitioners. Unresolved RBI/NPCI regulatory questions add substance. Not a global paradigm shift, but significant for payments infrastructure in one of the world's largest digital-payments markets.
Practice with real Payments data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Payments problems
