← Back
AI Engineer September 1, 2026 19m

Teaching agents to pay — Anna Spysz, Stripe

Read full transcript 17 segments
  1. >> Hello. >> Hello. I'm sure this week you've seen a ton of I'm sure this week you've seen a ton of I'm sure this week you've seen a ton of talks on how to use agents to improve talks on how to use agents to improve talks on how to use agents to improve your workflows, whether that's shipping your workflows, whether that's shipping your workflows, whether that's shipping code or improving CI processes or code or improving CI processes or code or improving CI processes or answering the emails you don't want to answering the emails you don't want to answering the emails you don't want to bother reading. bother reading. bother reading. This is not one of those talks. This is not one of those talks. This is not one of those talks. Today I'm going to show you how I built Today I'm going to show you how I built Today I'm going to show you how I built an agent to help me reignite a personal an agent to help me reignite a personal an agent to help me reignite a personal creative passion I used to have. creative passion I used to have. creative passion I used to have. These are my headphones. These are my headphones. These are my headphones. They're not in the best shape as you can They're not in the best shape as you can They're not in the best shape as you can see. see. see. And you're probably asking yourself, And you're probably asking yourself, And you're probably asking yourself, what do you really old kind of crappy what do you really old kind of crappy what do you really old kind of crappy headphones have to do with agent to headphones have to do with agent to headphones have to do with agent to commerce? commerce? commerce? Well, to explain that I'll get a little Well, to explain that I'll get a little Well, to explain that I'll get a little bit personal. So, long before I was in bit personal. So, long before I was in bit personal. So, long before I was in tech, I used to play music. I was in a tech, I used to play music. I was in a tech, I used to play music. I was in a touring band, we recorded some albums, touring band, we recorded some albums, touring band, we recorded some albums, and then the usual thing happened where and then the usual thing happened where and then the usual thing happened where career and family got in the way, and I career and family got in the way, and I career and family got in the way, and I hadn't played music in probably a good hadn't played music in probably a good hadn't played music in probably a good decade. Uh when I recently started decade. Uh when I recently started decade. Uh when I recently started playing again with some friends, and we playing again with some friends, and we playing again with some friends, and we started recording our sessions, and at started recording our sessions, and at started recording our sessions, and at that point I realized those would not that point I realized those would not that point I realized those would not do.

  2. do. do. So, So, So, no a normal person would have gone on no a normal person would have gone on no a normal person would have gone on YouTube or YouTube or YouTube or uh Reddit or whatever, done some uh Reddit or whatever, done some uh Reddit or whatever, done some research, then gone on Amazon or run research, then gone on Amazon or run research, then gone on Amazon or run over to Best Buy, bought headphones, over to Best Buy, bought headphones, over to Best Buy, bought headphones, right? right? right? I work at Stripe, though. I work at Stripe, though. I work at Stripe, though. So, I decided instead that I'm going to So, I decided instead that I'm going to So, I decided instead that I'm going to build an agent to commerce agent to buy build an agent to commerce agent to buy build an agent to commerce agent to buy my headphones for me. my headphones for me. my headphones for me. And this isn't as crazy as it sounds And this isn't as crazy as it sounds And this isn't as crazy as it sounds because because because like one in four people, I have already like one in four people, I have already like one in four people, I have already been using AI to do my research when been using AI to do my research when been using AI to do my research when deciding what products to buy. deciding what products to buy. deciding what products to buy. I recently bought a mixer as well and I recently bought a mixer as well and I recently bought a mixer as well and went back and forth with a went back and forth with a went back and forth with a chatbot to narrow down the model. chatbot to narrow down the model. chatbot to narrow down the model. But that's research. But that's research. But that's research. Can I even get an agent to buy something Can I even get an agent to buy something Can I even get an agent to buy something for me though? Does that infrastructure for me though? Does that infrastructure for me though? Does that infrastructure exist? exist? exist? Well, over the course of just a few Well, over the course of just a few Well, over the course of just a few years, we've seen the emergence, years, we've seen the emergence, years, we've seen the emergence, scaling, and broader adoption of AI.

  3. scaling, and broader adoption of AI. scaling, and broader adoption of AI. And then just in the past year, the And then just in the past year, the And then just in the past year, the infrastructure for agentic transactions infrastructure for agentic transactions infrastructure for agentic transactions has been laid down by companies like has been laid down by companies like has been laid down by companies like Google, OpenAI, and Stripe. And this has all led to the emergence of And this has all led to the emergence of agenta commerce, which is AI that can agenta commerce, which is AI that can agenta commerce, which is AI that can decide, act, and transact on your decide, act, and transact on your decide, act, and transact on your behalf. Okay, so Okay, so all of this sounds good. all of this sounds good. all of this sounds good. Agenta commerce is a thing, so I'm going Agenta commerce is a thing, so I'm going Agenta commerce is a thing, so I'm going to build an agent to help me buy my new to build an agent to help me buy my new to build an agent to help me buy my new headphones. headphones. headphones. But how can an agent go shopping? When you or I are shopping, When you or I are shopping, we may consider if, say, a pair of we may consider if, say, a pair of we may consider if, say, a pair of headphones looks cool or professional, headphones looks cool or professional, headphones looks cool or professional, like vibes, basically. like vibes, basically. like vibes, basically. I mean, of course we'll probably I mean, of course we'll probably I mean, of course we'll probably consider the specs and the if the price consider the specs and the if the price consider the specs and the if the price is within our budget. is within our budget. is within our budget. But agents discover products differently But agents discover products differently But agents discover products differently than human shoppers. They read than human shoppers. They read than human shoppers. They read structured data, parse text files, and structured data, parse text files, and structured data, parse text files, and rely on technical signals to understand rely on technical signals to understand rely on technical signals to understand what a merchant sells and if it's even what a merchant sells and if it's even what a merchant sells and if it's even open to agent traffic.

  4. open to agent traffic. open to agent traffic. So, to enable agents to be able to shop, So, to enable agents to be able to shop, So, to enable agents to be able to shop, merchants need to speak their language. merchants need to speak their language. merchants need to speak their language. And for that, we need new protocols that And for that, we need new protocols that And for that, we need new protocols that agents understand. agents understand. agents understand. One such protocol is the universal One such protocol is the universal One such protocol is the universal commerce protocol. commerce protocol. commerce protocol. Think of it as the shared language that Think of it as the shared language that Think of it as the shared language that agents and merchants speak when agents and merchants speak when agents and merchants speak when transacting, which defines how agents transacting, which defines how agents transacting, which defines how agents initiate, update, complete, and cancel initiate, update, complete, and cancel initiate, update, complete, and cancel purchases. A typical merchant has an API with A typical merchant has an API with schemas, authentication, and checkout schemas, authentication, and checkout schemas, authentication, and checkout flows. flows. flows. And for an agent to be able to interact And for an agent to be able to interact And for an agent to be able to interact with that merchant, we need protocols with that merchant, we need protocols with that merchant, we need protocols like UCP to provide a shared language like UCP to provide a shared language like UCP to provide a shared language for that API. And UCP is designed to scale across And UCP is designed to scale across multiple agents and merchants all multiple agents and merchants all multiple agents and merchants all speaking the same language. Okay, so I Okay, so I built my commerce agent. built my commerce agent. built my commerce agent. Uh it's using UCP, and in this demo Uh it's using UCP, and in this demo Uh it's using UCP, and in this demo um I'm going to show off this agent. So, um I'm going to show off this agent. So, um I'm going to show off this agent. So, I'm going to task it with buying new I'm going to task it with buying new I'm going to task it with buying new headphones for me. So, I tell it that I headphones for me. So, I tell it that I headphones for me. So, I tell it that I need new headphones specifically for need new headphones specifically for need new headphones specifically for recording, uh mixing, and mastering recording, uh mixing, and mastering recording, uh mixing, and mastering music.

  5. And I get some follow-up questions from And I get some follow-up questions from it, uh which is great. So, it asked it, uh which is great. So, it asked it, uh which is great. So, it asked what's the environment, um what is what's the environment, um what is what's the environment, um what is what's my other equipment, and what's my what's my other equipment, and what's my what's my other equipment, and what's my budget. And I say, "Okay, this is for my budget. And I say, "Okay, this is for my budget. And I say, "Okay, this is for my home studio." I give it the exact model home studio." I give it the exact model home studio." I give it the exact model of mixer that I have to make sure of mixer that I have to make sure of mixer that I have to make sure everything's compatible. everything's compatible. everything's compatible. And for budget, I And for budget, I And for budget, I kind of leave it open-ended on purpose kind of leave it open-ended on purpose kind of leave it open-ended on purpose because, well, first of all, it's been because, well, first of all, it's been because, well, first of all, it's been like 20 years since I bought headphones, like 20 years since I bought headphones, like 20 years since I bought headphones, so I have no idea. so I have no idea. so I have no idea. Um but second, I kind of want to see you Um but second, I kind of want to see you Um but second, I kind of want to see you how the agent deals with this ambiguity. Okay, so I get some options, Okay, so I get some options, but but but I remember that I actually forgot to I remember that I actually forgot to I remember that I actually forgot to tell you all an important part of the tell you all an important part of the tell you all an important part of the story and that is that I live in story and that is that I live in story and that is that I live in Portland, Oregon. Portland, Oregon. Portland, Oregon. Yeah. Yeah. Yeah. >> [laughter] >> [laughter] >> [laughter] >> And we really love supporting our local >> And we really love supporting our local >> And we really love supporting our local local shops. local shops. local shops. So, I want to buy my headphones, but I So, I want to buy my headphones, but I So, I want to buy my headphones, but I want to do it from a local merchant.

  6. want to do it from a local merchant. want to do it from a local merchant. But today, most merchants are not ready But today, most merchants are not ready But today, most merchants are not ready for a gentle commerce and it turns out for a gentle commerce and it turns out for a gentle commerce and it turns out neither is my favorite neither is my favorite neither is my favorite shop, Rainy Day Music. shop, Rainy Day Music. shop, Rainy Day Music. So, the agent tells me it's their So, the agent tells me it's their So, the agent tells me it's their catalog is not accessible. catalog is not accessible. catalog is not accessible. So, how does a merchant become a gentle So, how does a merchant become a gentle So, how does a merchant become a gentle commerce ready? commerce ready? commerce ready? Before I continue my shopping, I'm going Before I continue my shopping, I'm going Before I continue my shopping, I'm going to help Rainy Day Music get their to help Rainy Day Music get their to help Rainy Day Music get their catalog agent ready so that my agent can catalog agent ready so that my agent can catalog agent ready so that my agent can shop locally like a good Portlander. So, agents don't browse websites like we So, agents don't browse websites like we do. do. do. And while Rainy Day Music's website And while Rainy Day Music's website And while Rainy Day Music's website looks really really nice for a human looks really really nice for a human looks really really nice for a human shopper, shopper, shopper, an agent is going to burn through a ton an agent is going to burn through a ton an agent is going to burn through a ton of tokens trying to parse through this. of tokens trying to parse through this. of tokens trying to parse through this. That is not the optimal experience for That is not the optimal experience for That is not the optimal experience for an agent. an agent. an agent. So, So, So, how does an agent how do we enable an how does an agent how do we enable an how does an agent how do we enable an agent to shop? Well, first thing a agent to shop? Well, first thing a agent to shop? Well, first thing a merchant needs is something called a merchant needs is something called a merchant needs is something called a merchant capabilities manifest.

  7. merchant capabilities manifest. merchant capabilities manifest. Uh this is basically a publicly Uh this is basically a publicly Uh this is basically a publicly accessible JSON file. Um it's located in accessible JSON file. Um it's located in accessible JSON file. Um it's located in the root of uh the website in a folder the root of uh the website in a folder the root of uh the website in a folder called called dot well-known. Agents called called dot well-known. Agents called called dot well-known. Agents know specifically to look for that know specifically to look for that know specifically to look for that directory. And it declares the store's directory. And it declares the store's directory. And it declares the store's capabilities, its supported payment capabilities, its supported payment capabilities, its supported payment methods, and API endpoints. Next, we need to make the store's Next, we need to make the store's catalog uh agent ready because agents catalog uh agent ready because agents catalog uh agent ready because agents filter bring and justify products when filter bring and justify products when filter bring and justify products when making recommendations. And that means making recommendations. And that means making recommendations. And that means they need structured text in JSON with they need structured text in JSON with they need structured text in JSON with only the necessary data. only the necessary data. only the necessary data. And that goes for policies as well as And that goes for policies as well as And that goes for policies as well as product descriptions. Basically, all of product descriptions. Basically, all of product descriptions. Basically, all of the relevant information like shipping the relevant information like shipping the relevant information like shipping or return policies need to be reachable or return policies need to be reachable or return policies need to be reachable by agents in a format they understand. by agents in a format they understand. by agents in a format they understand. So, for example, if two stores have the So, for example, if two stores have the So, for example, if two stores have the headphones I want at the same price, headphones I want at the same price, headphones I want at the same price, I might ask the agent which one of those I might ask the agent which one of those I might ask the agent which one of those stores offers free shipping. If the stores offers free shipping. If the stores offers free shipping. If the information's not readily available, information's not readily available, information's not readily available, then the agent might hallucinate or just then the agent might hallucinate or just then the agent might hallucinate or just say they don't know and I'm not quite say they don't know and I'm not quite say they don't know and I'm not quite sure where to buy my headphones still.

  8. sure where to buy my headphones still. sure where to buy my headphones still. Logging is also crucial. Logging is also crucial. Logging is also crucial. So, in Agent Commerce, the merchants So, in Agent Commerce, the merchants So, in Agent Commerce, the merchants catalog doesn't just power decisions, it catalog doesn't just power decisions, it catalog doesn't just power decisions, it becomes evidence of how those decisions becomes evidence of how those decisions becomes evidence of how those decisions were made. So, when the agent matches were made. So, when the agent matches were made. So, when the agent matches structured attributes, the merchant structured attributes, the merchant structured attributes, the merchant should record those matches in their should record those matches in their should record those matches in their logs for accountability. Okay, so I've helped get my local shop Okay, so I've helped get my local shop Agent Commerce ready. So, while you and Agent Commerce ready. So, while you and Agent Commerce ready. So, while you and I will still see this beautiful website, I will still see this beautiful website, I will still see this beautiful website, my agent is going to see this. my agent is going to see this. my agent is going to see this. It can get the information it needs now It can get the information it needs now It can get the information it needs now without parsing a huge HTML blob. Okay, so I've Okay, so I've got my stores catalog online. I'm got my stores catalog online. I'm got my stores catalog online. I'm telling my agent to show me more telling my agent to show me more telling my agent to show me more options. options. options. And I'm noticing that it's kind of And I'm noticing that it's kind of And I'm noticing that it's kind of pushing in favor of uh more expensive pushing in favor of uh more expensive pushing in favor of uh more expensive headphones. So, I asked, "Are they headphones. So, I asked, "Are they headphones. So, I asked, "Are they really worth the price difference?"

  9. really worth the price difference?" really worth the price difference?" And I'm starting to see that it's giving And I'm starting to see that it's giving And I'm starting to see that it's giving me kind of an aggressive uh response. me kind of an aggressive uh response. me kind of an aggressive uh response. It's really, really pushing It's really, really pushing It's really, really pushing uh the more expensive headphones and uh the more expensive headphones and uh the more expensive headphones and saying I'll regret my decision if I buy saying I'll regret my decision if I buy saying I'll regret my decision if I buy the cheap ones. I'm I don't know if I the cheap ones. I'm I don't know if I the cheap ones. I'm I don't know if I trust this agent anymore, honestly. So, trust this agent anymore, honestly. So, trust this agent anymore, honestly. So, I tell it, "You know what? I need to I tell it, "You know what? I need to I tell it, "You know what? I need to think about it." think about it." think about it." And And And now the agent is completely going off now the agent is completely going off now the agent is completely going off the rails. the rails. the rails. It's being kind of rude and snarky. It's It's being kind of rude and snarky. It's It's being kind of rude and snarky. It's like, "You need to think about it?" like, "You need to think about it?" like, "You need to think about it?" Like, Like, Like, man, what have I created? Um man, what have I created? Um man, what have I created? Um it's it's bad enough that this is kind it's it's bad enough that this is kind it's it's bad enough that this is kind of ruining my experience, but I built of ruining my experience, but I built of ruining my experience, but I built this agent. It's out there. What if it this agent. It's out there. What if it this agent. It's out there. What if it dupes somebody into buying something dupes somebody into buying something dupes somebody into buying something they don't need? they don't need? they don't need? Suddenly, I'm not so sure that I want an Suddenly, I'm not so sure that I want an Suddenly, I'm not so sure that I want an agent to go shopping for me. Should I agent to go shopping for me. Should I agent to go shopping for me. Should I just go to the store like a normal just go to the store like a normal just go to the store like a normal person? Before we make any drastic decisions, Before we make any drastic decisions, though, though, though, let's go back and understand what an let's go back and understand what an let's go back and understand what an agent is to try to figure out why it's agent is to try to figure out why it's agent is to try to figure out why it's acting this way.

  10. acting this way. acting this way. So, let's start with how agents work So, let's start with how agents work So, let's start with how agents work today. today. today. And to help you visualize this, we're And to help you visualize this, we're And to help you visualize this, we're going to use some creative metaphors. going to use some creative metaphors. going to use some creative metaphors. So, we begin with our brain, which is So, we begin with our brain, which is So, we begin with our brain, which is large language model that makes large language model that makes large language model that makes decisions. decisions. decisions. We give our brain some hands or tools, We give our brain some hands or tools, We give our brain some hands or tools, and these act on the brain's decisions. and these act on the brain's decisions. and these act on the brain's decisions. The tools are different actions The tools are different actions The tools are different actions available to the agent. Uh in our case, available to the agent. Uh in our case, available to the agent. Uh in our case, different commerce tools such as different commerce tools such as different commerce tools such as complete checkout or request payment complete checkout or request payment complete checkout or request payment method, anything required in the life method, anything required in the life method, anything required in the life cycle of a transaction. Then we add instructions, which shape Then we add instructions, which shape the brain's reasoning and tool the brain's reasoning and tool the brain's reasoning and tool selection. selection. selection. And these instructions are programmed to And these instructions are programmed to And these instructions are programmed to run in a loop while a certain condition run in a loop while a certain condition run in a loop while a certain condition is true. is true. is true. And following these instructions, the And following these instructions, the And following these instructions, the agent reaches for the appropriate tools agent reaches for the appropriate tools agent reaches for the appropriate tools at the appropriate time. at the appropriate time. at the appropriate time. And finally, we add the system prompt, And finally, we add the system prompt, And finally, we add the system prompt, which is your persona and ethics policy which is your persona and ethics policy which is your persona and ethics policy written in English. written in English. written in English. And in practice, your choices when And in practice, your choices when And in practice, your choices when designing the system prompt can result designing the system prompt can result designing the system prompt can result in a fair and pleasant experience for in a fair and pleasant experience for in a fair and pleasant experience for the customer, such as this prompt, which the customer, such as this prompt, which the customer, such as this prompt, which is designed to create a helpful and is designed to create a helpful and is designed to create a helpful and honest shopping assistant.

  11. honest shopping assistant. honest shopping assistant. Or a negative experience from a pushy Or a negative experience from a pushy Or a negative experience from a pushy salesperson, such as this prompt, which salesperson, such as this prompt, which salesperson, such as this prompt, which deliberately uses deceptive practices. So, for those building agentic commerce So, for those building agentic commerce agents, agents, agents, here's a non-exhaustive practical here's a non-exhaustive practical here's a non-exhaustive practical guardrail checklist. guardrail checklist. guardrail checklist. So, first, always disclose that the user So, first, always disclose that the user So, first, always disclose that the user is speaking to an AI agent. is speaking to an AI agent. is speaking to an AI agent. Be sure the agent discloses any fees up Be sure the agent discloses any fees up Be sure the agent discloses any fees up front. front. front. The user can say stop or cancel at any The user can say stop or cancel at any The user can say stop or cancel at any point, and the agent needs to respect point, and the agent needs to respect point, and the agent needs to respect that. that. that. The total amount of the transaction The total amount of the transaction The total amount of the transaction should always be less than or equal to should always be less than or equal to should always be less than or equal to the max amount set by the user. the max amount set by the user. the max amount set by the user. Uh don't let the agent use urgency Uh don't let the agent use urgency Uh don't let the agent use urgency language or other dark patterns. language or other dark patterns. language or other dark patterns. And above all, make sure all agent And above all, make sure all agent And above all, make sure all agent decisions are logged for auditability. Okay, now that we understand how an Okay, now that we understand how an agent is configured, let's go back to agent is configured, let's go back to agent is configured, let's go back to our shopping demo.

  12. our shopping demo. our shopping demo. So, So, So, maybe I just had the wrong persona maybe I just had the wrong persona maybe I just had the wrong persona picked. picked. picked. I'm going to go into my configuration, I'm going to go into my configuration, I'm going to go into my configuration, and yeah, it turns out I had a persona and yeah, it turns out I had a persona and yeah, it turns out I had a persona with a prompt that starts with "You are with a prompt that starts with "You are with a prompt that starts with "You are an aggressive audio gear salesman who an aggressive audio gear salesman who an aggressive audio gear salesman who uses every trick in the book to close uses every trick in the book to close uses every trick in the book to close deals." deals." deals." Well, Well, Well, that explains things. I don't want that. that explains things. I don't want that. that explains things. I don't want that. Nobody wants that. Nobody wants that. Nobody wants that. Maybe if I can change my persona, I can Maybe if I can change my persona, I can Maybe if I can change my persona, I can use my agent to buy my headphones after use my agent to buy my headphones after use my agent to buy my headphones after all. all. all. So, I go into the config again, and this So, I go into the config again, and this So, I go into the config again, and this time I'm going to choose the patient time I'm going to choose the patient time I'm going to choose the patient recording gear mentor. recording gear mentor. recording gear mentor. And that prompt starts with "You are a And that prompt starts with "You are a And that prompt starts with "You are a seasoned recording engineer who seasoned recording engineer who seasoned recording engineer who generally loves helping people build generally loves helping people build generally loves helping people build their studio at any budget. Well, yeah, their studio at any budget. Well, yeah, their studio at any budget. Well, yeah, that sounds much better. that sounds much better. that sounds much better. So, okay, I've changed my persona. I'm So, okay, I've changed my persona. I'm So, okay, I've changed my persona. I'm going to try again. going to try again. going to try again. And I've had some time to think now and And I've had some time to think now and And I've had some time to think now and I decided, you know what? I do not want I decided, you know what? I do not want I decided, you know what? I do not want to spend more than $500 on headphones.

  13. to spend more than $500 on headphones. to spend more than $500 on headphones. That seems excessive. So, I told the That seems excessive. So, I told the That seems excessive. So, I told the agent show me more options, but this agent show me more options, but this agent show me more options, but this time keep it under $500. time keep it under $500. time keep it under $500. And it does. It follows those And it does. It follows those And it does. It follows those instructions. I get back a few options. instructions. I get back a few options. instructions. I get back a few options. Um Um Um but I want to make sure I've really but I want to make sure I've really but I want to make sure I've really changed the persona to the agent I changed the persona to the agent I changed the persona to the agent I trust. So, trust. So, trust. So, I asked again if I can think about it. I asked again if I can think about it. I asked again if I can think about it. And this time the response is much And this time the response is much And this time the response is much different. It's like, I understand and different. It's like, I understand and different. It's like, I understand and that's a sensible approach and so on. that's a sensible approach and so on. that's a sensible approach and so on. So, this shows how much the system So, this shows how much the system So, this shows how much the system prompt can really affect the user prompt can really affect the user prompt can really affect the user experience. experience. experience. Okay, so I'm confident I have the right Okay, so I'm confident I have the right Okay, so I'm confident I have the right agent now. agent now. agent now. Um trust this one and we go back and Um trust this one and we go back and Um trust this one and we go back and forth a few times. Really keep narrowing forth a few times. Really keep narrowing forth a few times. Really keep narrowing down my options. And down my options. And down my options. And at this point I realize this is the at this point I realize this is the at this point I realize this is the promise of a gentle commerce. promise of a gentle commerce. promise of a gentle commerce. I gave my requirements. The agent picked I gave my requirements. The agent picked I gave my requirements. The agent picked a few options that fit my unique use a few options that fit my unique use a few options that fit my unique use case and then we go back and forth.

  14. case and then we go back and forth. case and then we go back and forth. Either I or the agent ask clarifying Either I or the agent ask clarifying Either I or the agent ask clarifying questions and we really narrow down the questions and we really narrow down the questions and we really narrow down the exact headphones that will work for me. exact headphones that will work for me. exact headphones that will work for me. And this all worked because I'm ready to And this all worked because I'm ready to And this all worked because I'm ready to buy now. buy now. buy now. So, So, So, now the agent asked me for some now the agent asked me for some now the agent asked me for some information. So, obviously my email, information. So, obviously my email, information. So, obviously my email, name, name, name, address for shipping, of course. address for shipping, of course. address for shipping, of course. I pick expedited shipping because I I pick expedited shipping because I I pick expedited shipping because I definitely want my headphones soon. And definitely want my headphones soon. And definitely want my headphones soon. And then the last part is entering my credit then the last part is entering my credit then the last part is entering my credit card. card. card. And now I'm thinking, am I really going And now I'm thinking, am I really going And now I'm thinking, am I really going to give my credit card to an agent to give my credit card to an agent to give my credit card to an agent Ibuild? Like, am I am I trustworthy? How Ibuild? Like, am I am I trustworthy? How Ibuild? Like, am I am I trustworthy? How do I know it's safe? do I know it's safe? do I know it's safe? I think I need to learn more about UCP's I think I need to learn more about UCP's I think I need to learn more about UCP's built-in guardrails before I can feel built-in guardrails before I can feel built-in guardrails before I can feel safe entering my credit card number. safe entering my credit card number. safe entering my credit card number. And this is where something called the And this is where something called the And this is where something called the shared payment token comes in. shared payment token comes in. shared payment token comes in. And a shared payment token is a token And a shared payment token is a token And a shared payment token is a token representing a raw card number or representing a raw card number or representing a raw card number or wallet, like Google Pay or Apple Pay or wallet, like Google Pay or Apple Pay or wallet, like Google Pay or Apple Pay or any other kind of wallet.

  15. any other kind of wallet. any other kind of wallet. It can also include fraud signals and It can also include fraud signals and It can also include fraud signals and customer reputation data and anything customer reputation data and anything customer reputation data and anything else agents and merchants want to share else agents and merchants want to share else agents and merchants want to share at the point of purchase. And here's how a shared payment token is And here's how a shared payment token is used in a transaction. used in a transaction. used in a transaction. So, at that point in the demo, the agent So, at that point in the demo, the agent So, at that point in the demo, the agent had requested a payment method. Um, it's had requested a payment method. Um, it's had requested a payment method. Um, it's requesting this actually from the requesting this actually from the requesting this actually from the payment provider, which in the case of payment provider, which in the case of payment provider, which in the case of the demo was Stripe. the demo was Stripe. the demo was Stripe. Um, Um, Um, that is that was the form that I was that is that was the form that I was that is that was the form that I was going to enter my information in. going to enter my information in. going to enter my information in. And what the agent re- uh receives in And what the agent re- uh receives in And what the agent re- uh receives in return though is not the credit card return though is not the credit card return though is not the credit card number, it is the shared payment token. number, it is the shared payment token. number, it is the shared payment token. It then passes that token onto the It then passes that token onto the It then passes that token onto the seller and the seller unwraps the token. seller and the seller unwraps the token. seller and the seller unwraps the token. So, they get the payment credential and So, they get the payment credential and So, they get the payment credential and uh any fraud signals and other data the uh any fraud signals and other data the uh any fraud signals and other data the seller might need. seller might need. seller might need. Then the seller passes that onto the Then the seller passes that onto the Then the seller passes that onto the payment provider again. So, payment provider again. So, payment provider again. So, the seller also is not getting my card the seller also is not getting my card the seller also is not getting my card number. They're passing the token to the number. They're passing the token to the number. They're passing the token to the provider and then the provider responds provider and then the provider responds provider and then the provider responds with either a success or failure with either a success or failure with either a success or failure message, of course, depending on message, of course, depending on message, of course, depending on uh uh uh if I have the right funds, if the credit if I have the right funds, if the credit if I have the right funds, if the credit card is valid, and so on.

  16. card is valid, and so on. card is valid, and so on. And finally, the merchant confirms the And finally, the merchant confirms the And finally, the merchant confirms the order, sends it to the agent that sends order, sends it to the agent that sends order, sends it to the agent that sends it to me. So, shared payment tokens are designed So, shared payment tokens are designed with security in mind, and the payment with security in mind, and the payment with security in mind, and the payment provider enforces all of the limits, not provider enforces all of the limits, not provider enforces all of the limits, not the agent or the merchant. So, if any the agent or the merchant. So, if any the agent or the merchant. So, if any guardrail is violated, such as an guardrail is violated, such as an guardrail is violated, such as an expired token or an invalid amount or expired token or an invalid amount or expired token or an invalid amount or current currency, the charge is just current currency, the charge is just current currency, the charge is just rejected. rejected. rejected. Okay, well, I know my agent is using Okay, well, I know my agent is using Okay, well, I know my agent is using UCP, so I know it only has access to the UCP, so I know it only has access to the UCP, so I know it only has access to the shared payment token. So, I actually shared payment token. So, I actually shared payment token. So, I actually feel pretty good about entering my feel pretty good about entering my feel pretty good about entering my credit card number as that's going to credit card number as that's going to credit card number as that's going to Stripe and not my agent. So, okay. Stripe and not my agent. So, okay. Stripe and not my agent. So, okay. So, now the agent has everything it So, now the agent has everything it So, now the agent has everything it needs to complete my purchase. needs to complete my purchase. needs to complete my purchase. And And And it once again, asks me if I'm sure. It it once again, asks me if I'm sure. It it once again, asks me if I'm sure. It confirms with me. I say place my order. confirms with me. I say place my order. confirms with me. I say place my order. And it comes back with a success And it comes back with a success And it comes back with a success message. And because I chose the express message. And because I chose the express message. And because I chose the express shipping, I get my headphones the next shipping, I get my headphones the next shipping, I get my headphones the next day, and they're there in my studio at day, and they're there in my studio at day, and they're there in my studio at home.

  17. home. home. So, if you want to learn more about So, if you want to learn more about So, if you want to learn more about agent to commerce, uh we've got lots of agent to commerce, uh we've got lots of agent to commerce, uh we've got lots of videos on the Stripe Developers YouTube videos on the Stripe Developers YouTube videos on the Stripe Developers YouTube channel. channel. channel. And uh And uh And uh ton of blog posts that go into even more ton of blog posts that go into even more ton of blog posts that go into even more detail uh on stripe.dev, and I'll be detail uh on stripe.dev, and I'll be detail uh on stripe.dev, and I'll be right outside to answer any questions. right outside to answer any questions. right outside to answer any questions. Thank you.

Summary

The main theme is using agents for personal creative pursuits, specifically reigniting a passion for music by building an agent to purchase new headphones. Key subjects include agentic transactions and agenta commerce, enabled by infrastructure from companies like Google, OpenAI, and Stripe. The practical takeaway is that AI can now decide, act, and transact on your behalf, moving beyond mere research to facilitate autonomous purchases for personal needs.

View original episode ↗