Welcome to P3 Media’s AI Commerce Brief, your daily update on the AI and commerce stories shaping how companies build, sell, and grow. It’s Wednesday, September 9. Let’s get into it.
Meta is moving agentic commerce from product discovery into transaction execution.
The company launched Muse, a personal AI agent now rolling out in the United States on iOS, Android, and the web. Meta says Muse can work across apps, keep running after a user closes the app, and ask for approval before sensitive actions such as a purchase.
The commerce stack is notable. Stripe says US consumers can connect Link to Muse. At more than 1 million businesses that accept Link, Muse can use the shopper’s saved payment method. For other businesses, Link can issue a single-use virtual card scoped to the approved purchase. Meta says Shop Pay is coming soon.
Shopify also added Meta as an AI channel inside Agentic Storefronts. Shopify says products are shared with Meta by default through Shopify Catalog, while merchants can manage catalog and direct-checkout access and view Meta performance in agentic settings.
This gives agent-led shopping a mainstream test inside Meta’s consumer ecosystem. For merchants, the immediate work is less glamorous but important: review catalog quality, channel permissions, checkout settings, and how agent-referred orders are measured.
In marketplace news, Amazon says its India seller base has crossed 2 million, with more than 300,000 new sellers joining.
Amazon also reported a 7.5-times increase in year-over-year interactions with Seller Assistant. The share of advertisers using its AI-powered tools rose 77 percent, and small and midsize businesses generated 62 percent more ad creatives with those tools in the first quarter of 2026.
Those are Amazon’s figures, and the company has not established that AI caused the seller growth. Still, the direction is useful. AI is moving into everyday marketplace work, from listings and inventory planning to ad creative, alongside expansion of the seller base.
For Shopify operators, a smaller update may remove a persistent catalog headache.
Product variants can now store up to 20 barcodes. Shopify supports standard identifiers including UPC, EAN, ISBN, GTIN, and ASIN, plus custom codes. Teams can update them in bulk or through product CSV files.
One important caveat: sales channels still read the first barcode when they expect a single value.
That matters for merchants handling the same item across supplier packaging, private-label systems, stores, and marketplace listings. It should reduce spreadsheet-based mapping, but teams still need to choose the primary barcode carefully.
In creative AI, OpenAI released ChatGPT Images 2.5.
The company says the model improves detail, reference fidelity, and multi-turn editing. It also says generation latency is up to 50 percent lower than Images 2.0. Images 2.5 is rolling out across ChatGPT, ChatGPT Work, and Codex tiers. Two API models, Flare and Sunburst, are also available.
For commerce and marketing teams, the practical test is whether targeted edits can preserve products, layouts, and brand treatment through repeated revisions. OpenAI specifically positions the models for product imagery, campaign creative, and high-volume generation, but production teams should validate those claims against their own review standards.
And in global model competition, CISA, the NSA, and the FBI accused six China-based AI companies of industrial-scale knowledge distillation from US frontier models.
The agencies named DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun, and Z.AI. They allege the companies extracted billions of tokens across millions of requests since at least late 2024. China’s foreign ministry rejected the accusations as unfounded.
For model providers and cloud platforms, this raises the stakes around account controls, traffic monitoring, and API-abuse detection. For enterprise buyers, it adds another layer of policy and supply-chain risk to model selection.
What to watch: early merchant visibility inside Meta’s new channel, real checkout completion and dispute behavior for Muse, and whether the US advisory produces new platform controls or policy action.
That’s your AI Commerce Brief for today. Thanks for listening.
Meta is moving agentic commerce from product discovery into transaction execution.
The company launched Muse, a personal AI agent now rolling out in the United States on iOS, Android, and the web. Meta says Muse can work across apps, keep running after a user closes the app, and ask for approval before sensitive actions such as a purchase.
The commerce stack is notable. Stripe says US consumers can connect Link to Muse. At more than 1 million businesses that accept Link, Muse can use the shopper’s saved payment method. For other businesses, Link can issue a single-use virtual card scoped to the approved purchase. Meta says Shop Pay is coming soon.
Shopify also added Meta as an AI channel inside Agentic Storefronts. Shopify says products are shared with Meta by default through Shopify Catalog, while merchants can manage catalog and direct-checkout access and view Meta performance in agentic settings.
This gives agent-led shopping a mainstream test inside Meta’s consumer ecosystem. For merchants, the immediate work is less glamorous but important: review catalog quality, channel permissions, checkout settings, and how agent-referred orders are measured.
In marketplace news, Amazon says its India seller base has crossed 2 million, with more than 300,000 new sellers joining.
Amazon also reported a 7.5-times increase in year-over-year interactions with Seller Assistant. The share of advertisers using its AI-powered tools rose 77 percent, and small and midsize businesses generated 62 percent more ad creatives with those tools in the first quarter of 2026.
Those are Amazon’s figures, and the company has not established that AI caused the seller growth. Still, the direction is useful. AI is moving into everyday marketplace work, from listings and inventory planning to ad creative, alongside expansion of the seller base.
For Shopify operators, a smaller update may remove a persistent catalog headache.
Product variants can now store up to 20 barcodes. Shopify supports standard identifiers including UPC, EAN, ISBN, GTIN, and ASIN, plus custom codes. Teams can update them in bulk or through product CSV files.
One important caveat: sales channels still read the first barcode when they expect a single value.
That matters for merchants handling the same item across supplier packaging, private-label systems, stores, and marketplace listings. It should reduce spreadsheet-based mapping, but teams still need to choose the primary barcode carefully.
In creative AI, OpenAI released ChatGPT Images 2.5.
The company says the model improves detail, reference fidelity, and multi-turn editing. It also says generation latency is up to 50 percent lower than Images 2.0. Images 2.5 is rolling out across ChatGPT, ChatGPT Work, and Codex tiers. Two API models, Flare and Sunburst, are also available.
For commerce and marketing teams, the practical test is whether targeted edits can preserve products, layouts, and brand treatment through repeated revisions. OpenAI specifically positions the models for product imagery, campaign creative, and high-volume generation, but production teams should validate those claims against their own review standards.
And in global model competition, CISA, the NSA, and the FBI accused six China-based AI companies of industrial-scale knowledge distillation from US frontier models.
The agencies named DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun, and Z.AI. They allege the companies extracted billions of tokens across millions of requests since at least late 2024. China’s foreign ministry rejected the accusations as unfounded.
For model providers and cloud platforms, this raises the stakes around account controls, traffic monitoring, and API-abuse detection. For enterprise buyers, it adds another layer of policy and supply-chain risk to model selection.
What to watch: early merchant visibility inside Meta’s new channel, real checkout completion and dispute behavior for Muse, and whether the US advisory produces new platform controls or policy action.
That’s your AI Commerce Brief for today. Thanks for listening.
