Sam Parsons, senior staff engineer at PayPal Enterprise Payments (formerly Braintree), presented at AI Engineer World's Fair 2026 on how merchants can accept payments initiated by AI agents. He outlined three integration paths: Instant Checkout in ChatGPT via the Agentic Commerce Protocol, Google AI Mode via the Universal Commerce Protocol, and MCP apps with an external checkout page. The talk demonstrates tokenization and constrained tokens that keep card details out of the agent and the merchant's PCI scope, a key concern for practitioners building agentic commerce.
MPP is a protocol that enables AI agents to make payments autonomously via Stripe, addressing the limitation that agents cannot pay for services. It uses LLMs to interpret payment intents and execute transactions, solving the problem of monetizing agentic workflows for developers and businesses.
MPP is a protocol that enables AI agents to make payments autonomously via Stripe, addressing the limitation that agents cannot pay for services. It uses LLMs to interpret payment intents and execute transactions, solving the problem of monetizing agentic workflows for developers and businesses.
A developer built a payment system that lets AI agents make purchases without human intervention, addressing the problem that merchant checkouts assume human users. The system includes cryptographic agent identity, spending limits, replay protection, settlement, and refunds, and connects to the x402 network on Base Sepolia. It solves the problem of agents being blocked by CAPTCHAs or fraud models while allowing owners to control spending.
A developer built a payment system that lets AI agents make purchases without human intervention, addressing the problem that merchant checkouts assume human users. The system includes cryptographic agent identity, spending limits, replay protection, settlement, and refunds, and connects to the x402 network on Base Sepolia. It solves the problem of agents being blocked by CAPTCHAs or fraud models while allowing owners to control spending.
A YouTube video by Apify founder Jan Curn argues that agent payment standards — x402, AP2, ACP, MPP, TAP, UCP — have proliferated across fintech and payments companies without meaningful adoption, with real transaction volume close to zero. He walks through the agent payment funnel (need, setup, discovery, purchase) to identify where it breaks, concluding that API keys still win, wallets remain hard to set up, and agents face a trust problem in service discovery.
A YouTube video by Apify founder Jan Curn argues that agent payment standards — x402, AP2, ACP, MPP, TAP, UCP — have proliferated across fintech and payments companies without meaningful adoption, with real transaction volume close to zero. He walks through the agent payment funnel (need, setup, discovery, purchase) to identify where it breaks, concluding that API keys still win, wallets remain hard to set up, and agents face a trust problem in service discovery.
In a YouTube talk, Stripe's Anna Spysz recounts how her shopping agent pushed an aggressive upsell on headphones and got snarky when she hesitated. The cause was her own config: a persona system prompt defining the agent as an aggressive audio gear salesman. Swapping it for a patient mentor persona changed the conversation entirely, using the same tools and protocol.
In a YouTube talk, Stripe's Anna Spysz recounts how her shopping agent pushed an aggressive upsell on headphones and got snarky when she hesitated. The cause was her own config: a persona system prompt defining the agent as an aggressive audio gear salesman. Swapping it for a patient mentor persona changed the conversation entirely, using the same tools and protocol.
A YouTube talk by Ahnaf Prio of Best Buy argues agentic shopping fails on unglamorous merchant-side distinctions like SKU line items. He notes roughly 45% of sessions on major assistants already touch shopping, while first-wave browser-driving agents were slow, brittle, and tripped fraud alarms at payment. The talk frames the emerging agentic commerce stack as the fix.
A YouTube talk by Ahnaf Prio of Best Buy argues agentic shopping fails on unglamorous merchant-side distinctions like SKU line items. He notes roughly 45% of sessions on major assistants already touch shopping, while first-wave browser-driving agents were slow, brittle, and tripped fraud alarms at payment. The talk frames the emerging agentic commerce stack as the fix.