Agentic Commerce Comes of Age: How Walmart, Spotify, Zillow and Salesforce Are Reshaping Retail, Media, Real‑Estate and Enterprise

Introduction: the dawn of agentic commerce

2025 will be remembered as the year conversational AI stopped being a curiosity and started becoming a shopping channel. On 14 October 2025 Walmart announced that U.S. Walmart and Sam’s Club customers will soon be able to shop directly in ChatGPT using Instant Checkout, letting people restock groceries or buy gifts by talking to a chatbotmediapost.com. Spotify introduced its own app inside ChatGPT to let free and premium listeners create playlists or get song recommendations from an LLMrecommender-systems.com. Real‑estate giant Zillow became the first and only real‑estate app available in ChatGPThousingwire.com. And Salesforce, together with OpenAI and Stripe, announced an enterprise‑grade Agentic Commerce Protocol to embed instant checkout and conversational agents into CRM workflowspymnts.com.

These announcements are not isolated experiments. McKinsey calls agentic commerce a seismic shift that could generate US$1 trillion of orchestrated revenue in U.S. retail by 2030 and as much as US$3–5 trillion globallymckinsey.com. Conversational AI is already edging out traditional search: 44 % of users who tried AI‑powered search now say it is their preferred way to find informationmckinsey.com. The explosion of generative models, payment protocols and cross‑platform integrations suggests that commerce is moving beyond websites into intent‑driven, natural‑language experiences. Below, we examine the importance and impact of the four major partnerships—Walmart × OpenAI, Spotify × ChatGPT, Zillow × ChatGPT, and Salesforce × OpenAI/Stripe—across different verticals, and review the competitive landscape and implications.

1. Retail: Walmart × OpenAI, Etsy and Shopify

1.1 What the partnership does

• Instant checkout inside ChatGPT. OpenAI’s Instant Checkout allows U.S. ChatGPT users to buy a product directly in the chat. When someone asks for gift ideas or products, ChatGPT returns organic, unsponsored recommendationssearchengineland.com; if the item supports Instant Checkout, a “Buy” button lets the user confirm shipping and payment (Apple Pay, Google Pay, Stripe or credit card) without leaving the conversationtechcrunch.com. Merchants pay OpenAI a small fee; users pay nothingsearchengineland.com.
• Walmart integration. On 14 October 2025 Walmart announced that customers will soon be able to shop via ChatGPT using Instant Checkoutmediapost.com. CEO Doug McMillon said ecommerce’s “search bar and long list of responses” will give way to a multimedia, personalized and contextual AI experiencemediapost.com. Walmart’s system (powered by its Sparky AI assistant) will plan meals, restock kitchens and recommend giftsmediapost.com. Walmart’s internal adoption of AI has already reduced customer‑care resolution times by 40 % and shortened apparel production timelines by 18 weeksmediapost.com.
• Etsy and Shopify participation. Instant Checkout currently works for U.S. Etsy sellers and will soon include more than one million Shopify merchantsexplodingtopics.com. 74 % of Etsy’s gross merchandise sales come from the U.S., amounting to over $2 billion in Q2 2025, so opening this channel is significantexplodingtopics.com. More merchants will be onboarded before Black Fridayexplodingtopics.com.
• Agentic Commerce Protocol. OpenAI and Stripe co‑developed the Agentic Commerce Protocol (ACP), open‑sourcing it so any AI agent or payment processor can enable instant purchasesexplodingtopics.com. Merchants provide regularly refreshed product feeds and checkout APIs; AI agents can then present offers, secure payment tokens and share minimal dataopenai.com. The protocol is payment‑agnostic and aims to work with multiple processorsopenai.com.

1.2 Why it matters for retail

1. Discovery and purchase collapse into one interface. Traditional ecommerce requires browsing web pages, adding items to a cart and entering payment information. Instant Checkout eliminates these steps. Searches like “gifts for a ceramics lover” now produce product recommendations and the ability to buy immediatelytechcrunch.com. Research from Fortune shows that high‑consideration purchases—electronics, furniture, gifts—already generate 10 × higher conversion when handled by AI shopping agents compared with typical web experiencesfortune.com. This shift blurs the line between marketing and checkout, making conversational interfaces the new storefront.
2. Shift in traffic and power dynamics. ChatGPT reported 700 million weekly users, meaning the potential reach of Instant Checkout is enormousexplodingtopics.com. With 63 % of AI users already using chatbots for product researchexplodingtopics.com, agentic commerce could divert traffic away from Google search and retailer websites. TechCrunch notes that if product discovery happens inside AI chatbots, it could reduce the gatekeeping power of Google and Amazon, creating new power brokers for retailtechcrunch.com. At the same time, unsponsored ranking may challenge retail media models built on paid placementsearchengineland.com.
3. Opportunity for small sellers and new brands. Opening ChatGPT to Etsy and Shopify merchants gives millions of small businesses access to a massive audience. The Exploding Topics article notes that 5.4 million people operate Etsy stores, and 74 % of the platform’s sales come from the U.S.explodingtopics.com. When multi‑item carts arrive and other regions are addedsearchengineland.com, this channel could help niche brands compete against big-box retailers.
4. Data and brand control. Merchants remain the merchant of record, handle orders and fulfillment, and retain control over the customer relationshipsearchengineland.com. However, they must provide product feeds and checkout APIs and may face challenges optimizing for AI ranking (so‑called AI SEO)explodingtopics.com. Brands also worry that AI agents might suggest alternative or private‑label products, eroding brand loyaltyinvestopedia.com.
5. Competitive pressure on Amazon. Reuters notes that Walmart’s partnership aims to close the gap with Amazon and that ChatGPT referral traffic to Walmart in September already accounted for 15 % of referral traffic (though less than 1 % of overall traffic)reuters.com. Amazon responded by rolling out its own conversational assistant Rufus, which can answer product questions, compare items and track ordersaboutamazon.com. TechCrunch also highlights that Perplexity and Microsoft have similar shopping featurestechcrunch.com.

1.3 Implications for retailers and brands

The movement toward agentic commerce forces retailers to build agent‑ready infrastructure. The DAC Group advises businesses to prepare product feeds, checkout APIs and payment integration so that AI agents can discover and sell their productsdacgroup.com. Marketers must optimize their products for conversational discovery (non‑branded queries, natural language descriptions) and ensure their brands show up in AI responses. Brands should also guard against data misuse and ensure they maintain direct relationships with customers even when sales occur in third‑party chatbots. Finally, trust and transparency will be critical; systems like OpenAI’s ACP rely on user confirmation and tokenized paymentsopenai.com to mitigate fraud and maintain trust.

2. Entertainment: Spotify × ChatGPT

2.1 What the partnership does

• Personalized music in chat. On 6 October 2025 Spotify launched an app inside ChatGPT. Users mention “Spotify” in a prompt, authorize their Spotify account and can then ask for songs, albums, playlists or podcastsrecommender-systems.com. Free users draw on existing editorial and algorithmic lists like Discover Weekly or New Music Friday, while Premium users receive “fresh and fully personalized” selections based on mood, genre or topicrecommender-systems.com. The chat returns recommended tracks that link back to the Spotify apprecommender-systems.com. Spotify stresses the integration is opt‑in and that creators’ content will not be shared for AI trainingrecommender-systems.com.
• Framework for chat‑native apps. The integration is part of OpenAI’s apps in ChatGPT platform, which positions partner apps (Spotify, Booking.com, Canva, Figma, Zillow, etc.) as chat‑native servicesrecommender-systems.com. OpenAI provides an SDK and app directory; monetization models are still being explored.

2.2 Why it matters for music and media

1. Reimagining discovery. Spotify has long excelled at personalization through collaborative filtering and machine‑learning. The ChatGPT integration adds a natural‑language layer, letting listeners ask for very specific situations (“planning a rainy‑day sci‑fi binge with synthwave interludes”) and combining conversational intent with long‑term taste profilesrecommender-systems.com. This reduces the friction of searching through app menus and may increase time spent on Spotify’s ecosystem.
2. Free tier exposure. Both free and premium accounts can use the integrationrecommender-systems.com. Free users rely on existing editorial lists; the chat encourages them to upgrade to Premium for more personalization. The risk for Spotify is that ChatGPT becomes the entry point for discovery and dilutes in‑app engagement. VICE observed that the integration pulls users out of the Spotify app and could conflict with Spotify’s focus on keeping users engagedvice.com.
3. Trust and rights management. Spotify explicitly states that it will not share audio or video for AI training and that users can disconnect at any timerecommender-systems.com. This positions Spotify as protective of artists’ rights and may set a precedent for other media companies considering generative AI integrations.
4. Competitive landscape. Other streaming platforms such as Apple Music and YouTube Music have not yet announced similar integrations, but generative AI is creeping into media. The RS_c article notes that the challenge lies in combining historical taste data, conversational context and editorial priorsrecommender-systems.com. Spotify previously experimented with an AI DJ that uses an LLM for commentary; integrating an external LLM like ChatGPT changes the interface friction and may accelerate adoption of AI‑native music experiences.

2.3 Implications for media companies and artists

Music and media companies must decide whether to embed their services inside third‑party LLMs. Doing so provides new discovery channels but risks losing control over the user interface and data. They should negotiate clear terms about training data and user privacy and design experiences that drive users back to their own platforms for playback, thus preserving subscriptions and advertising revenue. Artists and labels may need to adapt metadata and descriptions to ensure their content is surfaced by AI agents; they should also monitor for biases or misrepresentations in AI‑generated recommendations.

3. Real Estate: Zillow × ChatGPT

3.1 What the integration does

• Natural‑language property search. On 6 October 2025 Zillow launched the only real‑estate app inside ChatGPThousingwire.com. Users can ask questions like “What can I afford nearby?” or “Show me homes with a big backyard,” and the Zillow app returns listings with photos, maps and pricinghousingwire.com. Users can then click through to Zillow to schedule tours, connect with agents or explore financinghousingwire.com. The integration currently supports rentals and homes for sale and will expand to include new construction and other property typeshousingwire.com.
• Compliance with listing data rules. Zillow says the chat experience uses its existing IDX‑compliant widget and does not transmit MLS data to unauthorized third partiesrealestatenews.com. However, some MLS leaders and realtors expressed concern that listing data displayed within ChatGPT might violate licensing rules; the National Association of Realtors advised MLSs to evaluate whether data is being transmitted to unauthorized partiesrealestatenews.com.

3.2 Why it matters for real estate

1. Top‑of‑funnel disruption. Real‑estate search has traditionally begun on Google or dedicated listing sites. By embedding Zillow into ChatGPT, property discovery now starts inside a conversational agent. Prospective buyers can ask open‑ended questions (“homes near good schools”) and receive curated results without wading through listings. This compresses the search, discovery and evaluation stages into a single conversation.
2. Importance of listing descriptions and metadata. ChatGPT relies heavily on text to interpret user intent and surface relevant homes. Real‑estate agent blogs note that features not mentioned in the listing description may not be surfacedjennifervaughan.com. Agents may need to write more detailed descriptions and add relevant keywords (e.g., energy‑efficient appliances, ADU potential) to ensure their properties appear in generative AI search results.
3. Data governance and compliance. The integration raises questions about who controls property data and how it is displayed. Zillow maintains that ChatGPT shows only their widget and does not store MLS datarealestatenews.com, but MLSs worry that listing information shown in a third‑party chat may violate IDX rulesrealestatenews.com. Real‑estate regulators may need to update policies to account for agentic commerce.
4. Fairness and bias. Because AI models learn from historical data, there is a risk that recommendations could perpetuate discriminatory patterns (e.g., steering certain buyers away from specific neighborhoods). Zillow says it worked with OpenAI to ensure data privacy and compliancerealestatenews.com. Agents and brokers must monitor for algorithmic bias and advocate for transparency.

3.3 Implications for the real‑estate industry

Real‑estate portals and brokerages should prepare for AI‑driven discovery by improving the quality and inclusiveness of their listing data, optimizing keywords, and ensuring compliance with data sharing rules. They could also develop their own AI assistants or integrate with existing LLMs to maintain brand presence. Brokers and agents may need to adapt their marketing strategies to appear in chat‑based searches and to provide additional value (e.g., hyper‑local expertise or video tours) that AI cannot replicate. Regulators should clarify how MLS data can be used by AI agents and ensure fairness and consumer protection.

4. Enterprise CRM and Commerce: Salesforce × OpenAI × Stripe

4.1 What the partnership does

• Agentforce 360 platform. Salesforce’s Agentforce 360 integrates conversational AI across its CRM, sales, service and marketing applicationsdigitalcommerce360.com. The October 14 rollout connects humans, agents and data in one platformdigitalcommerce360.com. It includes a chat interface, a unified data layer (Data 360) and integration with Customer 360 applicationsdigitalcommerce360.com. Slack serves as the collaboration interfacedigitalcommerce360.com. Salesforce reported that 12 000 customers are already using Agentforce 360 and documented improvements such as Reddit automating 46 % of customer support cases and Adecco Group shifting 51 % of candidate conversations outside working hoursdigitalcommerce360.com.
• Instant checkout for enterprises. The partnership with Stripe and OpenAI brings Instant Checkout to enterprise merchants. Salesforce’s Agentforce Commerce will integrate the Agentic Commerce Protocol so merchants can offer conversational checkout inside ChatGPT and Slackpymnts.com. The integration provides payment infrastructure and allows merchants to maintain control of the customer relationshippymnts.com.
• AI‑native workflows. OpenAI’s frontier models are embedded in Salesforce so that companies can query sales records, review customer conversations and generate Tableau visualizations directly within ChatGPTpymnts.com.

4.2 Why it matters for enterprises

1. Unifying communication, data and payments. Enterprise software is often fragmented across CRM, analytics, communication and commerce systems. Agentforce 360 brings these together with AI agents that can automate tasks, analyze data and initiate transactions. This agentic enterprise vision aims to augment rather than replace workersdigitalcommerce360.com. It could increase productivity by letting employees ask questions and complete tasks from one interface.
2. Scalable agentic commerce. The integration with Stripe uses the ACP to embed checkout in chat and ensure secure payment with minimal data sharingopenai.com. Businesses using Salesforce and Stripe can quickly enable agentic commerce; those using other processors can participate via shared payment tokensopenai.com. Salesforce notes that the unified system will accelerate revenue growth and provide new revenue streamssalesforce.com.
3. Competition with Microsoft and Oracle. Digital Commerce 360 reports that the Agentforce 360 rollout is intended to make Salesforce more competitive in the enterprise AI market, where Microsoft and Oracle are also expanding their offeringsdigitalcommerce360.com. Microsoft’s Copilot and Oracle’s generative AI services will likely integrate their own payment protocols, so Salesforce’s early adoption of ACP is strategic.
4. Governance and trust. Enterprises have stringent requirements for data security and compliance. The ACP and Salesforce’s governance tools provide an auditable trail and ensure that user authorization is explicitpymnts.com. This is critical for regulated industries (finance, healthcare) and may accelerate adoption.

4.3 Implications for enterprise software vendors

Enterprise vendors should evaluate whether to adopt open protocols like ACP or Google’s Agent Payments Protocol (AP2). AP2 uses cryptographically signed mandates to prove user intent and has the backing of more than 60 payments and tech partners, including Adobe, Mastercard, PayPal and Coinbasecmswire.com. AP2 aims to solve authorization, authenticity and accountability by generating verifiable credentials for each transactioncmswire.com. Microsoft, Oracle and other CRM providers will need to decide whether to support ACP, AP2, both or their own standards. Companies should also invest in data governance, identity management and consent frameworks so that AI agents can transact on behalf of employees and customers.

5. Competitive landscape and emerging protocols

Agentic commerce is not the exclusive domain of OpenAI and its partners. The ecosystem already includes:

Competitor / Platform	Key feature	Evidence
Amazon’s Rufus	Conversational shopping assistant trained on Amazon’s product catalog, reviews and Q&A; answers product questions, compares options and tracks orders for U.S. customersaboutamazon.com.	Amazon launched Rufus in 2024, giving U.S. customers a similar conversational buying experience.
Perplexity’s Buy with Pro	Shopping feature launched in November 2024; allows paying users to buy products directly from Perplexity if merchants have opted into its programexplodingtopics.com. Users receive free shipping even if merchants do not offer itexplodingtopics.com.	Demonstrates that agentic commerce is not limited to ChatGPT; Perplexity encourages adoption by bundling free shipping.
Google’s Agent Payments Protocol (AP2)	Open standard announced in September 2025 to let AI agents initiate and complete payments across platforms; uses cryptographically signed intent and cart mandates to verify user authorization and maintain an audit trailcmswire.com. Backed by more than 60 partners, including Mastercard, PayPal and Adobecmswire.com.	Provides an industry‑wide standard addressing authorization, authenticity and accountability.
Visa’s Intelligent Commerce	Payment system enabling AI agents to make pre‑authorized payments; allows users to set future payment instructions (e.g., ticket release) and leverages Visa’s networkexplodingtopics.com.	Shows payment processors are building their own agentic commerce protocols.
Microsoft Copilot Merchant Program	Allows merchants to create in‑chat storefronts; similar to Instant Checkouttechcrunch.com.	Highlights that large platforms (Microsoft, Amazon, Google) are building their own agentic ecosystems.

6. Importance and impact across verticals

6.1 For consumers

• Convenience and personalization. Agentic commerce shifts shopping from a self‑directed browsing activity to a conversational dialogue. Consumers can ask follow‑up questions and receive curated suggestions. Walmart’s CEO envisions a native AI experience that is multimedia, personalized and contextual, moving away from static listsmediapost.com. Spotify’s integration allows listeners to specify moods or topics and receive tailored playlistsrecommender-systems.com. Zillow’s app lets buyers see homes with specific features through natural languagehousingwire.com.
• Reduced friction. Eliminating the need to navigate websites or apps streamlines the purchase process. Instant Checkout allows confirmation of shipping and payment in one steptechcrunch.com. Real‑estate buyers can go from general curiosity to scheduling a home tour within the same chathousingwire.com. For complex purchases like electronics or furniture, AI agents provide guidance and comparison, boosting confidencefortune.com.
• Potential risks. Consumers must trust AI agents with payment details. OpenAI mitigates this by requiring explicit confirmation, using encrypted tokens and sharing only minimal dataexplodingtopics.com. Trust is a hurdle—Exploding Topics notes that only 8.5 % of users always trust AI overviewsexplodingtopics.com. Biases and fairness issues, especially in housing and lending, remain concernsjennifervaughan.com.

6.2 For businesses and brands

• New channels and data signals. Brands gain an additional “front door” to consumers through AI chatbots. However, they must optimize product feeds and descriptions for natural‑language queries and ensure brand identity is preserved within third‑party chat interfaces. Tools like AI SEO will become essentialexplodingtopics.com.
• Opportunity to differentiate. Early adopters like Walmart can position themselves as innovators and gain incremental traffic; UBS analysts believe Walmart’s move could drive more volume to its core consumables and grocery businessmediapost.com. Retailers who integrate early may see conversion lifts up to 10× compared with traditional websitesfortune.com.
• Risk of disintermediation. If AI agents control discovery and recommendation, brands may lose direct access to customers. This could erode loyalty and enable AI providers to favor their own products or partners. Merchants should invest in their own agents or negotiate favorable terms with AI platforms.
• Compliance and regulation. Companies must navigate data privacy, copyright and industry‑specific regulations (e.g., IDX rules for real estaterealestatenews.com). Payment protocols like AP2 and ACP offer frameworks for authorization and auditabilitycmswire.com. Adopting open standards can reduce fragmentation.

6.3 For the broader economy and society

McKinsey estimates that agentic commerce could contribute $3–5 trillion globally by 2030mckinsey.com. The Agentic Payments Protocol (AP2) and Agentic Commerce Protocol (ACP) could become as foundational as HTTP and TCP/IP were for the early web. Yet the concentration of power in a few AI platforms raises antitrust and competition questions. Regulators should monitor whether unsponsored ranking is truly neutral and whether AI providers leverage their position to favor their own products or data.

7. How businesses can prepare

1. Invest in agent‑ready infrastructure. Develop structured product feeds, descriptive metadata and checkout APIs that can be consumed by AI agentsdacgroup.com. Ensure inventory, pricing and availability are updated in real time.
2. Optimize for natural‑language discovery. Write product and listing descriptions that capture key features and attributes; incorporate synonyms, local context and relevant keywordsjennifervaughan.com. Monitor AI prompts and adjust content accordingly.
3. Adopt open protocols. Evaluate the Agentic Commerce Protocol (ACP) and Agent Payments Protocol (AP2). Align payment processors and CRM systems with these standards to ensure interoperability and verifiable trustcmswire.com.
4. Prioritize trust and ethics. Use encrypted payment tokens and explicit confirmation to build user confidenceexplodingtopics.com. Audit AI models for bias, particularly in sensitive verticals like housing and lending. Provide transparency on how recommendations are generated and allow users to opt out or adjust preferences.
5. Train employees and partners. Walmart has promoted AI literacy among its workforce and used ChatGPT Enterprise for employee trainingcorporate.walmart.com. Companies should invest in upskilling teams to work with AI agents, design conversational flows and interpret AI‑generated insights.
6. Experiment and iterate. The AI commerce ecosystem is evolving rapidly. Businesses should pilot agentic commerce in limited contexts (e.g., simple products or targeted audiences) and learn from early data before scaling. McKinsey suggests that companies must decide when and how to launch their own agents and whether to welcome agentic traffic, as these choices could be existentialmckinsey.com.

Conclusion

Agentic commerce is transitioning from experiment to reality. The partnerships between Walmart and OpenAI, Spotify and ChatGPT, Zillow and ChatGPT, and Salesforce with OpenAI and Stripe illustrate how conversational interfaces can unite discovery, recommendation and transaction in a single flow. Each vertical—retail, entertainment, real estate and enterprise—faces unique opportunities and challenges, yet common themes emerge: the importance of structured data, natural‑language optimization, open protocols, trust and early adoption.

Competition among tech giants will shape this landscape. Amazon’s Rufus, Perplexity’s Buy with Pro, Google’s AP2 and Visa’s Intelligent Commerce show that many players are vying to define the standards and capture the value of agentic commerce. Regulators and industry bodies must ensure that these systems remain open, fair and secure. For businesses and consumers alike, the coming years will be pivotal: those who adapt quickly could thrive in a world where shopping is as simple as having a conversation with an AI.