WEBVTT NOTE Sentence-level transcript of https://www.youtube.com/watch?v=qTZirYu9pr0 NOTE One cue per sentence. Cue ids are the line anchors on /transcripts/qTZirYu9pr0.html. A cue ends where the next begins, or 2 s after its last word. s1 00:00:01.309 --> 00:00:03.309 [music] s2 00:00:12.639 --> 00:00:17.760 Hello all, welcome to uh the agent e-commerce track and uh I'm Anil Lminti. s3 00:00:17.760 --> 00:00:20.080 I'm a senior solutions architect here at AWS. s4 00:00:20.080 --> 00:00:31.519 Uh I'm I'm here to talk to you today about how AWS is innovating and uh how you can build apps on uh AWS to support the agent e-commerce. s5 00:00:31.519 --> 00:00:33.600 So uh welcome to the session. s6 00:00:33.600 --> 00:00:38.480 Uh just to get you started, let me set the stage with something that you're very familiar with. s7 00:00:38.480 --> 00:00:44.640 Uh imagine that you are your organization is building uh a news portal like this, right? s8 00:00:44.640 --> 00:00:50.480 So you're all familiar with something where you're accessing the news content and then suddenly you hit a payw wall, right? s9 00:00:50.480 --> 00:01:02.239 So this is where uh you pull out your wallet or you try to uh figure out how to make the payments, set up your uh credentials, access keys in the sense that you you make a credit card transaction s10 00:01:02.239 --> 00:01:06.400 weekly, monthly or annual subscription and then get started to access the content. s11 00:01:06.400 --> 00:01:06.560 Right? s12 00:01:06.560 --> 00:01:08.880 So this is all the content that is behind a payw wall. s13 00:01:08.880 --> 00:01:18.479 But uh what we see now is that uh much of the traffic that is actually being uh sent to these uh uh portals now on the internet is all coming from bots. s14 00:01:18.479 --> 00:01:24.640 We see that we at a infliction point where uh the bot traffic is more than the human traffic right. s15 00:01:24.640 --> 00:01:31.840 So it's just uh actually in fact surpassed that and uh 95% of that bot traffic is coming from AI agents. s16 00:01:31.840 --> 00:01:47.600 So uh essentially we are also looking at the rise of autonomous agents right so where we all started using LLMs asking questions asking for summarization being able to uh get help with using them as co-pilots s17 00:01:47.600 --> 00:02:03.520 getting them to do agentic work to uh do multi-step tasks and now we're in the phase of autonomous agents where agents are using the reasoning powers of large language models to complete a task and completing a task means that uh it it has to go do whatever you're asking it to do. s18 00:02:03.520 --> 00:02:06.320 And that's kind of where we are in the uh journey. s19 00:02:06.320 --> 00:02:16.800 And we see that by 2027 about a billion agents will be running uh performing tasks and 60% of the enterprises will already be uh using agentic workflows. s20 00:02:16.800 --> 00:02:21.599 So uh what happens when agents hit these pay walls that uh we just saw? s21 00:02:21.599 --> 00:02:27.920 When agents hit the pay walls, they stall, they can't operate and you see those messages that hey I cannot access content. s22 00:02:27.920 --> 00:02:28.560 Uh right? s23 00:02:28.560 --> 00:02:30.879 So at that point humans get in the loop. s24 00:02:30.879 --> 00:02:38.560 They try to enter and put the credit card details or API keys do the transactions for the AI agents. s25 00:02:38.560 --> 00:02:40.480 But all of that is manual friction, right? s26 00:02:40.480 --> 00:02:42.720 So essentially bringing in a human in the loop. s27 00:02:42.720 --> 00:02:49.519 So autonom autonomous agents actually break [clears throat] at that point where the uh friction is now building up. s28 00:02:49.519 --> 00:02:54.640 So uh now sellers of the content have uh you know couple of options right. s29 00:02:54.640 --> 00:03:12.239 So block all the bot traffic but by blocking all the traffic they lose this AI powered discovery they miss this uh partnership licensing options and AI also now supports citations right so the responses so they lose all of that uh powered citations as well if they can't sell the content s30 00:03:12.239 --> 00:03:26.480 essentially they lose uh revenue generating options and if you allow the bots to access those uh uh the content what it means is that you know hundreds of thousands of bots or millions of bots could be hitting your uh uh infrastructure which means that the infrastructure s31 00:03:26.480 --> 00:03:37.840 costs will also raise and uh you need to be able to support all of that right so you also uh when you allow bots to access content you lose attribution the IP itself right so because content is now freely available s32 00:03:37.840 --> 00:03:54.400 so uh both these decisions are probably not not a good option they're not ideal so there should be another ideal option where uh you would want to have your AI agents being able to get and pay for the content that they are looking for and monetize on s33 00:03:54.400 --> 00:04:06.959 So uh now we look at the next phase of uh uh rise in autonomous agents where agents should be able to transact and make uh discover other agents resources and essentially make payments. s34 00:04:06.959 --> 00:04:07.200 Right? s35 00:04:07.200 --> 00:04:20.000 So this is the definition of agent e-commerce where AI agents can essentially discover uh you know it's a form of e-commerce where autonomous agents can uh discover independently and make those uh uh settlements and then access content. s36 00:04:20.000 --> 00:04:26.160 So let's look at uh the agent e-commerce the two sides of agent e-commerce the buy side and the sell side. s37 00:04:26.160 --> 00:04:35.040 So uh when we talk about the buy side the agents are making these transactions and on the sell side the sellers of the content are trying to monetize on the content. s38 00:04:35.040 --> 00:04:46.960 So on the buy side when you look at things AI agents want to access these premium paywalled content licensed content they want to be able to hold wallets which they do not have the option today and they want to be able to make these microtransactions s39 00:04:46.960 --> 00:04:48.960 you just heard in the prior talk as well. s40 00:04:48.960 --> 00:04:55.199 But enterprises when they come to this point they want more guardrails and they do not want agents to go on spending spree. s41 00:04:55.199 --> 00:04:56.000 Think about it right? s42 00:04:56.000 --> 00:05:03.280 Would you allow your AI agents to get handled on your wallets or credit cards to be able to do that transactions and where they could go rogue as well right? s43 00:05:03.280 --> 00:05:05.680 So that's what the buyer side is looking at. s44 00:05:05.680 --> 00:05:11.680 And on the seller side uh the there are again billions of transactions that be happening with these AI bots. s45 00:05:11.680 --> 00:05:20.960 So sellers really want to be able to understand what kinds of bots are operating uh what kinds of transactions they're making and uh really uh do this at the edge. s46 00:05:20.960 --> 00:05:26.400 The sellers don't want to change their entire infrastructure and origins where the content is sitting. s47 00:05:26.400 --> 00:05:29.680 They want to be able to do this at the edge without changing much of this. s48 00:05:29.680 --> 00:05:29.919 Right? s49 00:05:29.919 --> 00:05:39.919 So there is again uh one common thing here on the buyer side and the seller side which is a standardized approach or a protocol to be able to solve for this uh machine to machine payments at the edge. s50 00:05:39.919 --> 00:05:52.639 So uh bottom line buyers are saying that they want their agents to be able to pay for content uh and uh not have humans approving this and then the sellers are saying that they want to be able to earn from the AI traffic. s51 00:05:52.639 --> 00:06:03.199 So bottom line the subscription model is going to change with humans in the loop to becoming humans on the loop or out of the loop and that's kind of what we are building towards. s52 00:06:03.199 --> 00:06:14.960 uh the the traditional one-sizefits model does not work anymore because of the fact that uh again we look at that in the next slide where uh the transactions cost will not really work right all of this needs to happening s53 00:06:14.960 --> 00:06:28.479 at realtime speed and the paper use and paper execution is what the f future is going to look like so if you're a seller you would have come across this right so there is a 25 cent minimum transaction fees as well as s54 00:06:28.479 --> 00:06:38.319 2.5% on top of that and all of these microtransactions are uh you know in the in in like a cent subsend or you know micro cents is what we are calling them. s55 00:06:38.319 --> 00:06:44.960 So if you add like a 25 cents on top of that it's essentially like 250 times to what you know they are essentially paying for. s56 00:06:44.960 --> 00:06:57.840 So the all of this model does not work and uh while we are trying to solve for that a very brief history uh of this is every HTTP call essentially responds back uh you know there's a response for that you've seen 200 status codes s57 00:06:57.840 --> 00:07:08.960 404 uh you know and the 301 these are all like status codes that you're familiar with and then there is one status code which is 402 which has uh not been used it was reserved for payment required s58 00:07:08.960 --> 00:07:20.720 and now finally uh Coinbase has introduced this uh as uh transactions over 402 which is also called as X42 where they uh you know the protocol talks about how you can do machine to-achine s59 00:07:20.720 --> 00:07:34.800 transactions uh using this protocol right so we'll take a closer look at that but uh what happens within the protocol is uh you know if you look at this uh flowchart here a client makes a request to the server and then the server responds back with the payment required s60 00:07:34.800 --> 00:07:46.400 uh the client then figures out what is the payment method that it wants to operate and then it sends the payment authorization to the server the server then utilizes a facilitator to complete the verification s61 00:07:46.400 --> 00:08:03.360 and then also utilizes the same facilitator to complete the transaction and once the settlement is completed onchain essentially the server will then respond back with the content right so this is what's happening under the X42 protocol I thought I'll pick one of the protocols and just uh explain this to you but uh why this is compelling is s62 00:08:03.360 --> 00:08:18.960 uh you know essentially there is no protocol fees uh or the fees that a consumer is paying for uh you know these microcent transactions and the merchant is paying very nominal gas fees is uh again there is zero wait time this is happening at the speed of internet s63 00:08:18.960 --> 00:08:27.919 and uh there is no friction there is no API keys to set up no subscriptions and the payment is the essentially the uh credential to be able to get the content s64 00:08:27.919 --> 00:08:38.719 so there is no centralization it's uh x42 can be extended as well and you can implement it and there are no restrictions as well so some key milestones here are you know it was introduced s65 00:08:38.719 --> 00:08:57.519 last year uh May 2025 it explored is not part of the Linux Foundation under open governance and it's backed by Coinbase, AWS, Google, Stripe, Anthropic, Cloudflare and Circle right so many more folks in there that are supporting that organizations in there. s66 00:08:57.519 --> 00:09:09.040 So uh from Amazon we have also released agent core payments uh under the bedrock uh suite so where uh agents will make be able to make payments and we'll go into some of the details here. s67 00:09:09.040 --> 00:09:12.000 So let's talk about the buyer side here and what is involved right? s68 00:09:12.000 --> 00:09:17.920 So we uh we understood from the developers that they really want to be able to get this uh agents to have wallet support. s69 00:09:17.920 --> 00:09:28.240 They want to be able to have real-time settlement uh have the budget and guardrails which enterprises really want and observability throughout the stack where they would want to have s70 00:09:28.240 --> 00:09:31.839 uh the full stack trace of everything that's happening uh under the hood. s71 00:09:31.839 --> 00:09:42.720 So I'm excited to share with you that we've launched agent core payments and this is a service that allows AI agents to autonomously discover uh authorize and execute payments with a few lines of code. s72 00:09:42.720 --> 00:10:01.040 Uh now we've launched this in partnership with Coinbase and Stripe where you can bring wallets from Coinbase and Stripe preview to be able to do these operations and we'll go into some of the details but uh the core capabilities to start with are wallet support where you can bring the the wallets from Coinbase and Stripe. s73 00:10:01.040 --> 00:10:09.920 you're able to orchestrate the payments using payment connectors and uh today we support X42 with many more protocols to uh you know that are in the pipeline. s74 00:10:09.920 --> 00:10:12.399 The service is designed to be protocol agnostic. s75 00:10:12.399 --> 00:10:17.839 So as new protocols emerge, we are going to be adding the support for those protocols as well. s76 00:10:17.839 --> 00:10:32.720 And uh you know the the settlement is going to be instantaneous uh instant essentially and uh the payment limits can be set uh which is the most important thing that we spoke about where enterprises are looking to put some payment limits and guards on how these transactions can operate. s77 00:10:32.720 --> 00:10:42.480 So uh observability is builtin and uh essentially all of this uh operates with uh you know security as the uh layer that is operating uh the whole model right. s78 00:10:42.480 --> 00:10:47.680 So with that let's look at uh some of these uh details on how the payments limit can be set up right. s79 00:10:47.680 --> 00:10:59.839 So you can create payment sessions where you can set the programmatically set the maximum amount of uh uh value that can be used for transactions or you can also set expiry time in minutes. s80 00:10:59.839 --> 00:11:10.000 think where uh you are able to set that uh I can the agent can actually spend maybe $5 in 30 days or 60 days right so that's kind of the operation s81 00:11:10.000 --> 00:11:26.320 uh model that you can set with many more uh you know details that are available I'm only going over a few features but uh let's look at what happens on the buyer side when the user is asking an agent to make a particular uh you know requesting for resources right so agent s82 00:11:26.320 --> 00:11:31.839 completes the request by accessing tools MPPP servers other resources as well. s83 00:11:31.839 --> 00:11:42.160 So at that point of time if the agent uh is uh looking at you know it it finds that the there is a response from one of the tool calls or requests with a 402 s84 00:11:42.160 --> 00:11:55.120 agent core payments is going to handle the request uh to complete the transaction and then let the AI agent know that uh essentially the settlement happened and the AI uh agent will be able to respond back with the users. s85 00:11:55.120 --> 00:12:09.279 So in this process when the wallets are uh wallet support is imported the the the secret keys that you use to import the wallets actually are stored in a secure token wallet that is uh secured by KMS s86 00:12:09.279 --> 00:12:10.880 where you know that's there. s87 00:12:10.880 --> 00:12:14.160 So essentially the agent does not have access to the private keys. s88 00:12:14.160 --> 00:12:16.320 This is most important uh to note. s89 00:12:16.320 --> 00:12:25.839 And uh next thing is that uh agent core payments is also integrated uh through uh a gateway which is also part of uh which is another service that we have to mp5 s90 00:12:25.839 --> 00:12:27.440 your internal APIs. s91 00:12:27.440 --> 00:12:38.639 Uh through agent core gateway uh the agent core payments can get access to discovery service uh in coinbase where there are 10,000 plus endpoints that are available to transact s92 00:12:38.639 --> 00:12:43.200 and then u again there is a per session budget that we just discussed as well. s93 00:12:43.200 --> 00:13:03.440 So uh there is a decoupling of agent infrastructure and uh the payment infrastructure by design where the agent can operate it it in its own loop and whenever it sees the payment the payment uh connectors orchestration payment limits and integration with third party wallets can happen right so it's important to decouple them because again skills can be s94 00:13:03.440 --> 00:13:13.600 poisoned inputs for the agents can also be uh you know poisoned by inputs as well right so where malicious actors could uh try to do that so By decoupling s95 00:13:13.600 --> 00:13:26.639 and making this uh by design a agents can essentially have a secure path for these transactions and payments do not touch the you know undeterministic path but this is more on a deterministic s96 00:13:26.639 --> 00:13:28.320 uh uh layer as well. s97 00:13:28.320 --> 00:13:34.480 So why this is important is that uh again agents uh the code of the agents does not have to change. s98 00:13:34.480 --> 00:13:42.160 You can bring your own model frameworks and then the payment itself uh can flow through uh in the payment uh uh layer itself. s99 00:13:42.160 --> 00:13:51.920 So again the controls the policies pending controls can be outside of the payment stack itself and again it's this is built to be protocol agnostic. s100 00:13:51.920 --> 00:14:01.920 So this is uh one of the console screens where it shows how you can import uh the payment connector uh and it shows that you know you can select the the coinbase s101 00:14:01.920 --> 00:14:10.399 wallet and the stripe preview wallet from the console and uh this is a demo in action where we are showing how a secure resource can be accessed. s102 00:14:10.399 --> 00:14:22.320 Now in this case uh the AI agent is essentially making a you know discovering that there is a secure source the agent core payments is kicking in and then it's completing the transaction s103 00:14:22.320 --> 00:14:28.560 by utilizing the wallet that is already integrated and uh the transaction completes. s104 00:14:28.560 --> 00:14:31.760 Now uh this is on the buyer side. s105 00:14:31.760 --> 00:14:34.320 Now let's look at the seller side to understand what's happening. s106 00:14:34.320 --> 00:14:34.639 Right? s107 00:14:34.639 --> 00:14:38.000 So uh again there is a lot of bot activity that's happening. s108 00:14:38.000 --> 00:14:44.880 We have released uh under uh the uh AWS web application firewall a feature where we have bot detection in place. s109 00:14:44.880 --> 00:14:47.519 Today we detect over 650 different types of bots. s110 00:14:47.519 --> 00:14:53.120 Think of bots like perplexity bot, GPD bot, cloud bot, you know, again Google bots, right? s111 00:14:53.120 --> 00:14:55.120 So there are so many bots that are out there. s112 00:14:55.120 --> 00:14:58.800 So we're able to detect also understand the intent of these bots. s113 00:14:58.800 --> 00:15:00.720 So why are these bots accessing the content? s114 00:15:00.720 --> 00:15:03.440 Are they accessing the content to train their models? s115 00:15:03.440 --> 00:15:08.959 are they doing it because they have to respond back to an intent where they're uh responding for a rag search. s116 00:15:08.959 --> 00:15:10.720 So we're able to identify the intent. s117 00:15:10.720 --> 00:15:15.839 We're also able to verify the bots and identify them by a signature. s118 00:15:15.839 --> 00:15:18.720 So we are able to say hey this is a verified bot. s119 00:15:18.720 --> 00:15:29.680 So maybe you have uh built a relation with one of these organizations and these verification will allow you to have a different pricing for the organizations that are already verified. s120 00:15:29.680 --> 00:15:30.959 So we'll look at that in a second. s121 00:15:30.959 --> 00:15:36.320 So there's also real-time traffic analysis that allows uh more uh to be customized. s122 00:15:36.320 --> 00:15:41.839 And I'm also happy to share with you today that we announced VAF AI traffic monetization. s123 00:15:41.839 --> 00:15:52.639 This is a service that allows you to monetize uh based on the content that uh you know based on how you can measure, verify and monetize based on the AI traffic that is hitting your endpoints. s124 00:15:52.639 --> 00:16:02.399 Now if you might be familiar with CloudFront which is our content distribution network you can add a web application firewall at that point and essentially you can moni start monetizing s125 00:16:02.399 --> 00:16:07.759 uh right away and based on a few clicks uh you can do that again using infrastructure as code as well. s126 00:16:07.759 --> 00:16:16.079 Now I also spoke about a gateway service that allows you to expose your AI endpoints uh that are internal use and mcpify them. s127 00:16:16.079 --> 00:16:18.800 So the same web application firewalls can be used there. s128 00:16:18.800 --> 00:16:22.959 So your internal APIs can be MCPI and then you can start monetizing as well. s129 00:16:22.959 --> 00:16:24.959 So what happens during monetization? s130 00:16:24.959 --> 00:16:29.759 The AI agent AI bot essentially requests for some content. s131 00:16:29.759 --> 00:16:34.160 The bot context uh understands what kinds of bots is detecting it. s132 00:16:34.160 --> 00:16:35.440 It's able to detect the bot. s133 00:16:35.440 --> 00:16:43.680 It's able to categorize and understand the intent of the bot as we discussed earlier and verify uh and check what kind of bot is available. s134 00:16:43.680 --> 00:16:49.680 So then we are able to monetize uh using the X42 and the publishers get paid as well. s135 00:16:49.680 --> 00:16:54.880 So important to note is that again there is no SDK change no changes at the origin origin. s136 00:16:54.880 --> 00:17:01.519 Publishers keep 100% of the revenue as well and uh again there is no transaction fees or subscription fees. s137 00:17:01.519 --> 00:17:07.199 So this supports X42 and we are adding support for more uh protocols as well. s138 00:17:07.199 --> 00:17:10.240 U a few dimensions on how you can start monetizing. s139 00:17:10.240 --> 00:17:12.720 Think uh you have separate paths. s140 00:17:12.720 --> 00:17:24.000 So a slash uh uh blog in this case can be charging for a different rate than a slash research or maybe an API endpoint itself and uh you you know the identity of these bots. s141 00:17:24.000 --> 00:17:36.240 Again, if you make uh some kind of a relationship with the uh bots uh companies, organizations, maybe you make a relation with Anthropic, then you can essentially have a different pricing for those bots versus different s142 00:17:36.240 --> 00:17:38.000 uh unverified bots, right? s143 00:17:38.000 --> 00:17:39.600 So, uh think of that option. s144 00:17:39.600 --> 00:17:42.320 And then you can also set different pricing for intent as well. s145 00:17:42.320 --> 00:17:51.200 If somebody's coming here, if a bot is accessing the content for uh again training, you can charge a different rate than what it's doing for a search as well. s146 00:17:51.200 --> 00:17:53.200 So again, these are different VAF rules. s147 00:17:53.200 --> 00:17:57.760 they can be uh in a combination of end or or then you can access that. s148 00:17:57.760 --> 00:18:10.160 So this is how the reimagine flow would look like uh where you're allowing the AI agents or you know essentially verified bots and unverified bots to have different pricing and humans to have different pricing. s149 00:18:10.160 --> 00:18:12.960 Some cases you want to have humans to access the content freely. s150 00:18:12.960 --> 00:18:17.600 Some cases again the humans could be charged where the bots could be charged differently as well. s151 00:18:17.600 --> 00:18:17.840 Right? s152 00:18:17.840 --> 00:18:20.640 So this is how uh you know you can reimagine the price. s153 00:18:20.640 --> 00:18:34.960 So again there is some uh dashboards that show the revenue numbers and how uh you know you're uh able to aggregate by different uh bots and figure out what kind of revenue model you want to operate and it also shows what is the path s154 00:18:34.960 --> 00:18:46.160 uh that's being accessed by these bots right so uh again what is currently uh everyone using agent e-commerce for they're using agent commerce to uh run um again LLM inference s155 00:18:46.160 --> 00:18:56.160 getting uh compute web scraping uh they're uh creating research search agents to be able to uh you know serve the requests and uh agent to agent as well. s156 00:18:56.160 --> 00:18:58.799 We see MCPs also being monetized now. s157 00:18:58.799 --> 00:19:12.240 Uh again this is the last 12 months of uh traffic uh from um again what we are seeing on uh Coinbase agentic market u where you're seeing that a $50 million volume transaction happened over s158 00:19:12.240 --> 00:19:13.919 170 million transactions. s159 00:19:13.919 --> 00:19:22.559 The average settlement time is 200 milliseconds on base uh with about a tenth of a cent as cost per transaction. s160 00:19:22.559 --> 00:19:34.799 So I spoke to you about agent uh agent core payments which is uh one of the you know parts of the bigger ecosystem agent uh bed agent core uh where you can essentially bring your own model. s161 00:19:34.799 --> 00:19:40.640 You can bring your own framework and uh you can start building AI agents. s162 00:19:40.640 --> 00:19:43.120 You can add context by adding memory. s163 00:19:43.120 --> 00:19:45.840 You can uh bring again your own managed knowledge bases. s164 00:19:45.840 --> 00:19:48.080 You can add web search capabilities to the agents. s165 00:19:48.080 --> 00:19:56.960 you can mcpify your internal APIs and then you can uh have many more features like being able to run evaluation on how your agents are performing. s166 00:19:56.960 --> 00:20:11.919 So again you can uh use runtime which is a bedrock uh agent core runtime where you can bring your own agentic uh uh application and serve uh at scale and every request will have uh its own isolated s167 00:20:11.919 --> 00:20:15.440 uh micro virtual machine that is running to serve the requests. s168 00:20:15.440 --> 00:20:17.919 So that's it from my side here today. s169 00:20:17.919 --> 00:20:22.244 Uh thank you and uh hope you have a nice day. s170 00:20:22.244 --> 00:20:24.244 [applause] s171 00:20:38.619 --> 00:20:40.619 [music]