A PayPal Enterprise Payments (formerly Braintree) talk by Sam Parsons on three ways merchants can accept payments initiated by AI agents: Instant Checkout in ChatGPT using the Agentic Commerce Protocol with a merchant-built MCP app, Google AI Mode using the Universal Commerce Protocol and Google Pay with merchant endpoints, and an MCP app with an external checkout page. It includes demos and covers how tokenization keeps card details away from the agent and the merchant's PCI scope, and how tokens are constrained by merchant, amount, currency, and expiry. # Channel Presentation Observations ## Source Scope - **Video:** "How AI Agents Pay: Checkout in ChatGPT and Google AI Mode — Sam Parsons, PayPal" - **Available inputs:** Title and transcript only. No curated channel metadata was supplied, so no channel name, description, or other identity context is included here. ## Explicit Self-Description These are claims the speaker makes about himself and his organization. They are not independently verified. - **Speaker role:** "I'm from PayPal. Um, and I work on our Agentic Commerce area. Um, my name is Sam Parsons, senior staff engineer here." - **Organization identity:** The speaker explains a rebrand: "you may have heard of Braintree before... now it's PayPal enterprise payments." - **Business scale:** "So, we handle payments for really large corporations. You know, think of the biggest ones that you can think of. Huge volume, huge scale." - **Stated takeaway:** "PayPal Enterprise Payments is building the bridges for merchants to be able to accept payments from many different places." - **Self-framing of maturity:** "we're kind of early days here" and "this is in a fast evolving area." ## Intended Audience The transcript points to a technically oriented audience of merchants and integration developers. - **Merchant-facing framing:** Repeated references to what "the merchant" must build, control, or receive, such as "the merchant gets to control the look, the feel" and "from the merchant's perspective, they do a couple of things." - **Developer-facing detail:** The speaker shows real code ("I'm showing you the real code for... instant checkout with OpenAI"), describes endpoints, a `well-known` JSON file, and an MCP server "complete checkout tool." - **Event context:** "We also have a booth over there in the back of the room," and "I'm going to be there for the next hour or so," indicating a live conference audience. ## Claimed Scope The speaker describes the talk as covering **three paths** for accepting agentic payments: 1. **ChatGPT Instant Checkout** using the Agentic Commerce Protocol, delivered through an MCP app. 2. **Google AI Mode** using the Universal Commerce Protocol, with Google Pay tokenizing to PayPal. 3. **MCP apps with external checkout**, a human-in-the-loop path where checkout opens in an external page. The speaker also says the protocols are "very similar" and "almost identical" in their specs. This is a comparative claim made by the speaker. ## Demonstrated Presentation Patterns These patterns are observed within this single transcript. - **Live walkthrough:** The speaker narrates a demo of hotel booking in ChatGPT, then a Claude connector example. He pauses, fast-forwards, and "hit[s] play" during the recording. - **Concept explanation for non-specialists:** He defines tokenization for a general audience: "tokenization means where the user exchanges a real payment credential that's transactable... and instead they exchange that for a token." - **Trade-off framing:** He contrasts paths directly, such as: "there is a little bit more friction that's introduced in this mode because you're not doing like the actual checkout inside of your agent." - **Vendor-neutral framing in places:** He notes MCP apps work with "Goose or other ones," and mentions Google and ChatGPT alongside Claude. - **Acknowledged uncertainty:** He says some features are "not supported in most of the major ones yet" and "coming soon." ## Promotional and Call-to-Action Elements - Directs viewers to documentation via a QR code and to a sign-up process. - Invites questions in person at a booth. - The overall message is oriented toward encouraging merchants to adopt PayPal Enterprise Payments integrations. ## Limits of This Observation - This document describes only what this one video shows. It does not establish channel-wide patterns, recurring formats, or the channel's full audience. - The transcript contains apparent transcription artifacts (for example, "aentic" and "Brainree"), so exact wording is taken as rendered and should not be treated as authoritative spelling. - No timestamps are provided in the source, so none are cited. - The speaker's private motivations are not inferred. # Delivery Observation — Sam Parsons, PayPal ## Overall Communication Style The delivery is a **walkthrough-style technical talk**: verbal framing of a concept, followed by live demo narration, then integration detail. The speaker explicitly structures the talk for the audience ("we're going to talk about a couple of things, three different paths"), signaling a deliberate roadmap habit. > "So, what we're going to talk about here is a couple of things, three different paths." ## Tone - **Conversational and informal**, with frequent fillers and self-corrections: *"um," "you know," "kind of," "sorry that fast forward through there," "Apologies about that."* - **Directly audience-addressing**, using inclusive and second-person framing: > "I wanted to show you what you could really do today which is you know like this." > "So, I'm going to be there for the next hour or so and would love to talk to any of you who have any questions." ## Recurring Vocabulary and Phrases Several phrases appear repeatedly and look like stable verbal habits: - **"Kind of"** as a softener/hedge — very frequent: *"kind of maybe a human in the loop," "kind of a general standard," "Let me kind of walk through what that flow looks like," "kind of demos for you."* - **"So" as a sentence opener and pivot word** — appears dozens of times, often multiple times per segment: *"So first of all," "So that's what's going to happen right here," "So next I'm going to walk through."* - **"Walk through"** as the preferred verb for explanation: *"I'm going to walk through what that flow looks like," "Now I'm going to walk through what Google AI mode."* - **"Behind the scenes"** for internal mechanics: *"behind the scenes, that'll be tokenized," "what's kind of happening behind the scenes here," "using communication behind the scenes."* - **Merchant-centric framing** — "merchant gets to control," "from the merchant's perspective," "complete control as a merchant" recurs across all three paths. ## Teaching Habits - **Roadmapping before detail**: announces the three paths up front and restates the takeaway ("the key takeaway that I want you to have is that PayPal Enterprise Payments is building the bridges"). - **Naming and rebranding clarification**: proactively resolves terminology confusion early: > "You may have heard of Brainree before and Brainree is PayPal's large enterprise payments... if you know it as Brainree, uh, it's the same thing." - **Comparison as an explanatory device**: contrasts the OpenAI and Google protocols ("if you go look at the specs, you're going to find out, oh, they are almost identical") and contrasts control vs. friction across integration options. - **Plain-language definition mid-demo**, stopping the demo to define technical terms: > "So if you're kind of not familiar with payments, tokenization means where the user exchanges a real payment credential... for a token that can then be passed around without increasing PCI scope." - **Live demo narration with visible imperfection handling**: the speaker narrates what's on screen step-by-step and acknowledges glitches in real time (*"Hit play," "Apologies about that"*). ## Structural / Communication Patterns - **Enumeration habit**: numbered steps and lists recur ("that's step three," "step five," "two main key integration points," "a couple of key pieces here. One is..."). - **Trade-off framing**: repeatedly pairs benefits with costs — external checkout gives "a lot more control, but there is a little bit more friction." - **Forward-looking hedge**: positions the material as early-stage to manage expectations — *"we're kind of early days here," "a fast evolving area... you're going to see way more over the coming months."* ## Uncertainty Notes - The filler density ("um," "you know," "kind of") could reflect live-demo nervousness rather than a stable default style; a single talk is limited evidence. - The claim that "merchant-centric framing" is a deliberate emphasis is based on repetition within this talk only — repetition is clear, but attribution to habit versus topic necessity is not determinable from this evidence. - The apology moments ("Sorry," "Apologies about that") appear only during the demo segment and may be situational rather than a pattern. # Observations: "How AI Agents Pay: Checkout in ChatGPT and Google AI Mode" **Speaker:** Sam Parsons, identified in the transcript as a senior staff engineer at PayPal working on agentic commerce. **Organization context:** PayPal Enterprise Payments (formerly Braintree). The speaker says the two names refer to the same business. **Scope:** How merchants accept payments initiated by AI agents, with or without a human in the loop. --- ## 1. Core Thesis The speaker's stated "key takeaway": > PayPal Enterprise Payments is building the bridges for merchants to be able to accept payments from many different places. He presents this as a **fast-evolving area** and says he is showing "a flavor" of the options, with "way more" expected "over the coming months and year." --- ## 2. Protocol Landscape | Protocol / Standard | Attributed Origin (as stated) | Speaker's Characterization | |---|---|---| | Agentic Commerce Protocol (ACP) | OpenAI | Used for Instant Checkout in ChatGPT | | Universal Commerce Protocol (UCP) | Google, with "some others" | Used in Google AI Mode | - The speaker says the two protocols are **"very similar"** and that their specs are **"almost identical."** - He also says, "we're kind of early days here." - **MCP apps** are described as a separate, general-purpose path, supported by multiple agents (ChatGPT, Claude, and "Goose or other ones"). --- ## 3. Path One: Instant Checkout in ChatGPT (ACP) ### 3.1 Mechanism as Described - A merchant builds an **MCP app**: an MCP server with embedded UI (HTML, JavaScript) that the host agent renders. - The merchant controls discovery and product presentation within the app. - When the user chooses to buy, the MCP app asks the host (ChatGPT) to start a checkout. - The **host app** collects payment details and performs **tokenization** with PayPal. - The token is returned to ChatGPT and then passed to the merchant. ### 3.2 Speaker's Explanation of Tokenization > Tokenization means where the user exchanges a real payment credential that's transactable, say your credit card number and instead they exchange that for a token that can then be passed around without increasing PCI scope. He also says the merchant "doesn't have to do that" tokenization step. ### 3.3 Integration Points Named 1. **Client-side:** JavaScript within the MCP app widget that triggers the checkout experience. 2. **Server-side:** A call to the merchant's MCP server with a "complete checkout" tool, which delivers the token. The merchant then processes it "as usual" as a PayPal Enterprise Payments token. ### 3.4 Token Constraints (Stated) Tokens are described as: - Bound to a specific merchant - Capped at a maximum dollar amount - Bound to a currency - Time-limited by expiration Transactions are attributed to their source, so they appear in PayPal Enterprise Payments as ChatGPT transactions. --- ## 4. Path Two: Google AI Mode (UCP) ### 4.1 Mechanism as Described 1. The user shops in AI Mode. 2. The agent creates a **session** against the merchant on the user's behalf, using UCP. 3. At checkout, **Google Pay** tokenizes the payment with PayPal Enterprise Payments. 4. The token is returned to the merchant. 5. The merchant completes the checkout and PayPal performs the **charge**. The speaker says this is "actually a very similar model" to the ChatGPT path. ### 4.2 Key Difference Stated by the Speaker > The merchant doesn't create an MCP app... There is no like UI that you have to build. You have to stand up endpoints essentially to integrate with this. ### 4.3 Integration Points Named - **Discovery configuration:** A well-known JSON file for UCP that specifies processing through Braintree/PayPal Enterprise Payments, along with a merchant ID and a client key. The speaker says the client key "is safe to pass through in the public." - **Endpoint:** A call from UCP checkout session carrying a session ID and a "complete" action. The merchant receives a token representing the credit card behind the Google Pay payment. --- ## 5. Path Three: MCP Apps with External Checkout ### 5.1 Mechanism as Described - The MCP app widget shows a product card with a checkout button. - Clicking checkout opens an **external page** where the merchant controls the entire flow and accepts "any payment method that they want." - A session links the external checkout to the MCP app, which **updates automatically** on completion via inter-app communication. ### 5.2 Trade-offs Explicitly Acknowledged > There is a little bit more friction that's introduced in this mode because you're not doing the actual checkout inside of your agent. The speaker frames this path as offering **more merchant control** in exchange for that friction. ### 5.3 Forward-Looking Claim (Qualified) - MCP apps has a feature called **elicitation**, which the speaker says is "not supported in most of the major ones yet." - He says it could allow "a little bit more seamlessly" requesting external validation in an iframe, and that it is "coming soon." --- ## 6. Shared Post-Checkout Claims The speaker states that across all paths, the merchant's downstream operations are unchanged: > Settlements, refunds, chargebacks, compliance, all of that remains the same and is supported through PayPal enterprise payments in general. He describes the merchant's experience as "a really seamless integration" and says the merchant receives tokens that "look like every token they normally process." --- ## 7. Qualifications and Tensions Within the Talk - **"Universal" vs. "early days":** The protocol names suggest broad standards, but the speaker repeatedly qualifies the area as early and rapidly changing. - **Merchant control vs. host control:** The speaker says the merchant controls the MCP app's UI. However, in the ChatGPT Instant Checkout path, he says the checkout experience is *not* generated by the merchant: "This is provided from the instant checkout experience." Control is therefore split between merchant (discovery and presentation) and host (checkout UI and tokenization). - **"Seamless" vs. "friction":** He calls the Instant Checkout integration "really seamless" for merchants, yet describes the external-checkout MCP path as adding friction. These are presented as trade-offs, not contradictions, but the framing differs by path. - **Tokenization actor:** In the ChatGPT path, the host app performs tokenization. In the Google path, Google Pay tokenizes. The speaker presents both as tokenizing with PayPal, but the entity performing the step differs. - **Similarity claim:** He says the two protocols are "almost identical" and the flows are "very similar," while also noting different merchant integration requirements (MCP app UI build versus standing up endpoints). - **Demo caveats:** He notes the demo required fast-forwarding and that the checkout panel was not merchant-generated. He also apologizes for a delay, and the demo used a test credit card number. --- ## 8. Points Not Addressed in the Talk - Specific fee, pricing, or commercial terms. - Fraud, risk, or dispute outcomes specific to agent-initiated transactions beyond the statement that chargeback and compliance processes "remain the same." - Whether a human confirmation step is required in each path. The title and opening frame the topic as "maybe a human in the loop, maybe not," but the talk does not state which paths guarantee human approval. - Availability timelines, beyond "coming soon" for elicitation and "over the coming months and year" for general evolution. - Governance or ownership of the protocols beyond attribution to OpenAI and Google "with some others." --- ## 9. Source Notes - Speaker identity is taken from the transcript's self-introduction and the title. No separate curated metadata was supplied. - The transcript contains apparent automatic-transcription artifacts (for example, "aentic" for "agentic," "Brainree" and "Braintree" used interchangeably, and "Jat GPT" for ChatGPT). These are noted here and were not silently corrected in the attributed claims. - Demo details and on-screen content (QR codes, UI behavior) are described verbally by the speaker and are not independently verified in the transcript. # Attribution Observation: "How AI Agents Pay: Checkout in ChatGPT and Google AI Mode" **Video ID:** 01a15ca6-ebb7-7346-bdf5-3a2684aab6f1 **Source basis:** Transcript and title only. No curated metadata was supplied. --- ## 1. Speakers and Roles ### 1.1 Identified Speaker - **Name:** Sam Parsons - **Stated affiliation:** PayPal, as "senior staff engineer" - **Stated area:** PayPal's "Agentic Commerce" area - **Role in video:** Sole presenter. Delivers the talk and a screen-based demonstration. ### 1.2 Other Voices - **Audience:** Not identified. The presenter refers to possible questions, but no audience questions appear in the transcript. - **Music:** A music cue appears at the end of the transcript. No performer is identified. ### 1.3 Organizations Named by the Speaker - **PayPal Enterprise Payments:** Described as the presenter's employer and the provider of the payment services discussed. The presenter notes that the company was formerly known as **Braintree**, and says the two names refer to the same thing. - **OpenAI:** Described as having created the Agentic Commerce Protocol, which is "in ChatGPT." - **Google:** Described as having partnered with others to create the Universal Commerce Protocol. Also referenced as **Google AI Mode** and **Google Pay**. - **Anthropic / Claude:** Referenced only as an agent that renders MCP apps as "connectors." No claim is made about Anthropic beyond that. - **Goose:** Named as another agent that supports MCP apps. --- ## 2. Promotional Content > **⚠ Promotional flag:** The entire transcript is presented by a PayPal employee about PayPal's product offerings. Treat all statements about PayPal capabilities as **vendor-originated**, not independent. Specific promotional elements: - **Product pitch:** The presenter states the "key takeaway" is that PayPal Enterprise Payments "is building the bridges for merchants." - **Calls to action:** - A QR code linked to "our docs" for signing up and setup. - An invitation to visit "a booth over there in the back of the room" for the next hour or so. - **Scale claims:** The presenter says PayPal handles payments "for really large corporations" with "huge volume, huge scale." No figures are given. - **Forward-looking claims:** The presenter says the space "you're going to see way more" of "over the coming months and year," and describes an upcoming MCP apps feature called "elicitation" as "coming soon." These are vendor predictions. --- ## 3. Referenced Sources and Protocols ### 3.1 Named Standards and Technologies | Reference | How the speaker describes it | |---|---| | Agentic Commerce Protocol | Created by OpenAI; used for Instant Checkout in ChatGPT | | Universal Commerce Protocol (UCP) | Created by Google with partners; used in Google AI Mode | | MCP (Model Context Protocol) and MCP apps | Server plus embedded UI (HTML, JavaScript) rendered inside agents | | Google Pay | Performs tokenization in the Google AI Mode flow | | PCI scope | Referenced as the reason tokenization is used | ### 3.2 Third-Party Claims Made by the Speaker - The OpenAI and Google protocols are "very similar," and the presenter says that "if you go look at the specs... they are almost identical." The presenter does not name the specs. - Settlements, refunds, chargebacks, and compliance "remain the same" for merchants. This is a PayPal-side assertion. - The presenter describes how tokens are constrained (merchant-bound, maximum amount, currency, expiration). These are stated as design properties of PayPal's system. - Google's "well-known JSON file" is described as the place where merchants declare processing settings. The presenter does not cite a document. ### 3.3 Documentation Links - A QR code and an unnamed docs link are referenced. The transcript does not include URLs. **No specific documentation sources are verifiable from the transcript.** --- ## 4. Demonstrations The presenter describes two live or recorded walkthroughs: 1. **ChatGPT Instant Checkout**: A travel/hotel booking flow, including a test credit card entry. 2. **MCP app with external checkout**: A product card in a Claude connector that opens an external checkout page and then updates automatically. > **Note on recording:** The presenter says "sorry that fast forward through there" and "Hit play," which indicates at least part of the demo was pre-recorded or edited. The transcript does not say which portions. Some demo content is shown on screen and is **not present in the transcript**. Claims about what appeared on screen rest on the presenter's narration alone. --- ## 5. Availability Limits | Channel | Status | |---|---| | Speaker identity | Stated in transcript; not independently verified | | Slides or on-screen text | Not available in transcript | | Demo footage | Referenced; not available as text | | Links and QR code targets | Referenced; URLs not in transcript | | Q&A | Not captured | | Comment availability | Not supplied in source material | | Timestamps | Not supplied in source material | --- ## 6. Transcript Quality Notes These artifacts appear in the transcript. Recorded terms are normalized in this document: - "aentic" and "Aente" → *agentic* - "Brainree" → *Braintree* - "chat GBT" / "Jat GPT" → *ChatGPT* - "price payments" → likely *PayPal payments* - "Gentic" → *Agentic* Some sentences are incomplete or cut off, notably the description of the Google well-known file and the instructions for the complete-checkout endpoint. Okay. So, I'm going to talk to you about aentic commerce um with PayPal enterprise payments. So, this is about how do we accept payments from agents kind of maybe a human in the loop, maybe not um for merchants. So, I'm from PayPal. Um, and I work on our Aente Commerce area. Um, my name is Sam Parsons, senior staff engineer here. Um, so first of all, you may have heard of Brainree before and Brainree is PayPal's large enterprise payments. Was that Brainree Heart? Brainree. Yes. Uh, so but now it's PayPal enterprise payments. So I'm going to say PayPal enterprise payments, but if you know it as Brainree, uh, it's the same thing. So, we handle payments for really large corporations. You know, think of the biggest ones that you can think of. Huge volume, huge scale. So, what we're going to talk about here is a couple of things, three different paths. And really, the the key takeaway that I want you to have is that PayPal Enterprise Payments is building the bridges for merchants to be able to accept payments from many different places. And so, I'm going to give you a flavor of some of them, but really this this is in a fast evolving area. that you're going to see way more um over the coming months and and year. So, first of all, we're going to talk about a genic commerce protocol. So, this is uh OpenAI has created that and that'll be in chat GPT. You can use this universal commerce protocol. Google partnered with some others and created that protocol and these are very similar. So, if you go look at the specs, you're going to find out, oh, they are almost identical. Um, and again, you know, we're we're kind of early days here, but also we're going to demo for you another path towards accepting agentic payments, human in the loop using just MCP apps and and external checkout. Okay, so let's dive in. Um, so first of all, this is instant checkout and chat GPT using aentic commerce protocol. So, PayPal Enterprise Payments built an integration to a Gentic commerce protocol that allows you to start taking payments in this manner. So, there's a couple different flavors. The one I'm going to walk through uses an MCP app. So, an MCP app is like an MCP server, but you embed inside of that app. You embed UI. So, it has HTML, JavaScript, and things like that. And those then can get rendered inside of different agents. So, Jat GPT can render those. You can add an app. You can render them in Claude. You call that a connector. Same thing. It's an MCP app. And this extension many different agents out there, you know, Goose or other ones also support. So, this is kind of a a general standard. Instant checkout allows us within chat GPT to trigger like an in chat purchase of something. Okay. So let me kind of walk through what that flow looks like. So first of all, the user will have their intent. So they're searching in chat GPT. They'd like to buy something and move to the next part. There's discovery there. So the merchant using an MCP app can show to the user, here's things that you may want to purchase. They get to control that experience. So it's inside of say chat GPT, but the merchant gets to control it. They get to control the look, the feel. they get to control what's showing up and interact there. Then when the user decides, I'd like to go to the next step and buy this. They can trigger this checkout. So that's where the MCP app talks to its kind of host chat GPT and says, I'd like to purchase something that uses a gentic commerce protocol. behind the scenes, the user would enter in maybe a credit card or maybe they had that stored and move to the next part that will get tokenized against PayPal systems, PayPal enterprise payments. And finally, when we get to the end of that, they'll complete that and send that across to the merchant who then charges that. So, we kind of look at this big life cycle. At the end of it, from the merchants's perspective, they do a couple of things as far as bridging that gap. They're pro providing these MCP app experience to the user, but they're also providing \[clears throat\] they're receiving these tokens that come from ultimately PayPal enterprise payments. And that looks like every token they normally process. So this is a really seamless integration for them at the end of it. And on the back of it, settlements, refunds, chargebacks, compliance, all of that remains the same and is supported through PayPal enterprise payments in general. So right now we're just going to do a a walkthrough of searching here in chat GBT. This will give you that instant checkout experience. So this one here is like a travel app. So I'd like to buy some um looking for some hotels. Um here's some different options. And I'll just pause here. So you can see this UI here is not a chat GPT UI. This is controlled by the merchant. merchant gets to specify that they can have their interface for what makes sense depending on what they're selling. Um, sorry that fast forward through there. Hit play. And so then we continue here. When they're ready, they go to proceed to checkout. This now is pulling up the chat GBT instant checkout experience inside of there. So the the merchant isn't actually generating this. This is provided from the instant checkout experience. Apologies about that. And from there, they start entering in their credentials. So, they're going to enter in some credit card details. It's just a test credit card number that they're going to enter. And behind the scenes, that'll be tokenized with PayPal. So if you're kind of not familiar with payments, tokenization means where the user exchanges a real payment credential that's transactable, say your credit card number and instead they exchange that for a token that can then be passed around without increasing PCI scope and you know dangerous things happening. So that's what's going to happen right here behind the scenes is a tokenization will occur and that's the merchant doesn't have to do that. This is like within the host the host app is going to do that tokenization send it to PayPal enterprise payments and then we give back the token to chat GPT and that's safe to pass around. So this is what's happening right now. Tokenization is occurring and then it's going to be sent to the merchant. The merchant receives that and says okay now I transact with that token securely. And now you see they've come back to the MCP app. This has been updated. Their booking's been confirmed. And so you can see from a merchants's perspective, they have all control of what that UI looks like and it's pretty seamless for the the user. What are the integration points here? So the first integration point is inside of that MCP app widget. So this I'm showing you the real code for um for do this instant checkout with open AAI. You need to have this JavaScript code that triggers that checkout experience. So that's provided within the context of your MCP app. And then the next place where the merchant picks this up is they receive a call to their MCP server with this complete checkout tool and they get that token and then they can proceed as usual with their processing of you know a typical PayPal enterprise payments token. So it's really pretty straightforward as far as that goes. Now there's still obviously the work of creating an MCP app within that experience. Also, there's um links here to the QR code if you want to like follow and look at the docs for how you would set this up. And what's kind of happening behind the scenes here is these tokens are constrained in a number of ways. One of them is with the merchant so that this is can only be processed on a given merchant. It has a max amount of dollars that are applied. It has a currency. It also has an expiration. So, you can't just use this forever. And all of those transactions when processed they get attributed to the source where they came from. So this would show up within PayPal enterprise payments as a chat GPT um transaction that occurred. Um so next I'm going to walk through what Google AI mode and how we're meeting merchants there. So the flow looks a little bit different when you're processing transactions on Google AI mode. So first of all the user might be shopping in AI mode that you know want to buy you know say a pair of running shoes or something like that behind the scenes using the universal commerce protocol the agent will create a session against that merchant for the user. So that kind of exists as an entity belonging to the merchant and and it uses that. Then when the user says I now want to check out with this the tokenization happens but here this is using Google pay. So Google pay would tokenize with PayPal enterprise payments. And so that's step three and then give that token back to the merchant. And you kind of see this is actually a very similar model. Then at the completion step they provide that to the merchant. And at step five we do the charge. So the merchant comes along and says here's my token. I would like to process this transaction. here's all the information about it and the same sorts of things like this again settlement refunds chargeback risk compliance this all is just standard operating procedure as far as the merchants perspective but they receive this token from a different path this came in through Google AI mode one of the things to note here is in Google AI mode the merchant doesn't create an MCP app they don't create an experience their integration with Google AI mode is more providing here are my products, here is my endpoints for receiving the payments, the complete checkout payment um and so forth. There is no like UI that you have to build. You have to stand up endpoints essentially to to integrate with this. Um so you can see here one of them is specific specifying in your um well-known JSON file for universal commerce protocol and you specify some things that tell us or tell Google pay this is needs to be processed on brainree or PayPal and price payments and here's your merchant ID and here's your client key and that's safe to pass through in the public. Then Google will take that and when the final end of it is you'll get a call to this endpoint from UCP checkout session with that session ID and a complete and you then get to process that and what you'll receive in there is this um token representing the credit card behind that Google Pay payment. So again two main key integration points um as far as receiving those payments. The next one I'm going to show you here is about MCP apps and just kind of opening up your mind about what are the other things that you could do with MCP apps. This one gives the merchant really full control over the whole experience and we're going to work in things that are not necessarily Google or ChatGpt, but there's a lot of different um agents that already support MCP apps. So, a couple of key pieces here. One is that there's the MCP app widget. So this will in our example here will show like a product card and you'll get a checkout button. Then there'll be a handoff um that will open up the checkout into an external page. So then you can complete that there and then come back and get it automatically updated within your MCP app. So that gives you a lot more control, but there is a little bit more friction that's in introduced in this mode because you're not you're not doing like the actual checkout inside of your agent. So let's um walk through this one. So here we're going to search for some things uh that we'd like to buy. And this loads up an MCP app. And you see this is using cloud and this is a connector. So MCP apps show up as connectors or apps depending on you know the container. So here's this is the MCP app. The merchant would design this. They add the things that the user now is picking what they want and when they click checkout this is now going to pop up an external link and you can tie these together with a session as a merchant. So that's what's happening here. This opens up that session. They can then can complete things with any payment method that they want. You have complete control as a merchant over everything. But when you complete that, notice how this MCP app updates automatically. So, we're using communication behind the scenes to allow us to know exactly that it was completed and the status of that order and everything. So, that again just kind of demos for you what you can do with an MCP app and kind of what's on the table. And there's a lot of new things um coming along here. MCP apps has a new feature. It's not supported in most of the major ones yet that is called elicitation and so you can do that maybe a little bit more seamlessly as far as requesting that external validation in an iframe or something. So that's coming soon. Um I wanted to show you what you could really do today which is you know like this. This kind of gives you a little bit of an over overview of like what are the different options um that we offer merchants today and kind of what are some of the trade-offs. So like if you want to be in chat GPT or you want to be on Google um there's options for you or if you want more control as a merchant over the whole experience you also have um some options there too. And um yeah so that's um that's the main thing that I wanted to share with you. Uh there's a link here on this QR code if you want to look at our docs find out more about how to sign up and how to use that. And if you have any questions, um, I'd love to hear that. I don't know if we can maybe do questions here. Maybe just come talk to me afterwards. We also have a booth over there in the back of the room. So, I'm going to be there for the next hour or so and would love to talk to any of you who have any questions about uh what we offer. Thank you. \>\> \[music\]