Why the GS1 Resolver Will Become the Source of Truth for AI Answer Engines
Search is changing shape. More people now ask ChatGPT, Claude, Perplexity, and Google's AI answers to recommend a product, check an ingredient, or confirm where something was made, and fewer people scroll a page of blue links. A new layer of software sits between your product and the shopper, the answer engine. Soon AI shopping agents will not just describe products, they will compare and buy them on a person's behalf.
That raises a hard question for every CPG brand. When an AI decides what to say about your product, where does it get its facts? Today it stitches them together from marketplace listings, retailer pages, and scraped descriptions, and it often gets them wrong. This article explains why the GS1 Digital Link resolver is set to become the highest level of product authority for answer engines, what that means for brands preparing for GS1 Sunrise 2027, and how to set your resolver up so the machines reading your product reach for your data first.
What is answer engine optimisation, and why should product brands care?
Answer engine optimisation (AEO) is the work of making sure AI systems like ChatGPT, Claude, Perplexity, and Google AI answers represent your product accurately and cite your data when they respond. For product brands it matters because these engines increasingly sit between the shopper and the shelf, and they favour sources that are structured and verifiable.
An answer engine does not rank ten pages and let you pick. It interprets the question, gathers candidate sources, and writes a single answer from whatever it judges clearest and safest to repeat. That changes what wins. Structure and specifics beat volume of links.
The data backs this up. Around 65% of pages cited in AI answers use structured data, and pages with proper schema markup have roughly a 2.5 times higher chance of showing up in AI-generated answers than pages without it. Structured data is the current bar, and a GTIN-anchored resolver is the next rung above it. Unverifiable claims rarely get repeated, because the engine cannot check them.
For a product brand, the fact the engine is trying to verify is your product data. Its ingredients, its origin, its certifications, its claims. Being legible to machines is also becoming a sales channel in its own right. About 91% of retail IT leaders name AI the top technology they plan to implement by 2026, and live agentic purchases already run inside ChatGPT, Google AI Mode, and Copilot. The brand whose data is easiest to trust is the brand the machine repeats and, before long, the one it buys.
What is a GS1 Digital Link resolver?
A GS1 Digital Link resolver is a web service that connects a product's GTIN to the online resources tied to it. When a person or a machine reads a GS1 Digital Link QR code, the resolver decides what to return based on who is asking, so one code can serve a consumer page, a compliance document, or structured data for a system.
A GS1 Digital Link expresses the GTIN as a web address. The identifier on the pack and the link to the product's data become the same thing. The resolver is the routing layer that sits behind that link.
One point matters for accuracy. A resolver is not what makes a barcode work at checkout. A point-of-sale scanner reads the GTIN straight from the Digital Link structure and needs no internet connection to do it. The resolver handles everything beyond the till: consumer content, traceability data, compliance responses, and, increasingly, queries from AI agents. As James puts it, same code, different answers depending on who is asking. The GS1 Digital Link standard defines around 60 link types, and those link types are what let a resolver hand a machine a clean, structured response instead of a marketing page.
Why is the resolver the highest level of product authority for AI?
The resolver is the only endpoint that ties a product's globally unique identity to data the brand controls and keeps current, served in a machine-readable form. Scraped listings and duplicate product pages cannot match that mix of unique identity, brand control, and structure, which is exactly what an answer engine needs to trust a source.
Picture product authority as a pyramid. At the base sit scraped web pages and marketplace listings. They are inconsistent, often stale, and carry no guaranteed link to a single product identity. Above them sits schema markup on a product page, which is better but still page-level and easily contradicted across a dozen URLs for the same item. Higher still sit retailer and GDSN data pools, which are trusted but built for business-to-business exchange rather than the consumer or the agent.
At the top sits the brand's GS1 Digital Link resolver. It is anchored to the GTIN, so there is no ambiguity about which product the data describes. It is controlled by the brand, so the facts are the brand's own. It stays current, because the code on the pack never changes while the data behind it can be updated any time. And it can serve a machine-readable answer built for the requester. The vocabulary bridge already exists, because schema.org accepts a GS1 Digital Link URL as a valid product identifier.
Be clear about timing. This shift is coming, not fully here. Link types and resolvers start paying real dividends for AI as brands roll out 2D barcodes through 2027. The foundation is being laid now, which is exactly why the data layer decision belongs on the table today.
What happens when AI cannot find authoritative product data?
When an answer engine cannot find a trusted source, it fills the gap with whatever it can scrape, and it sometimes states the wrong ingredient, origin, or claim with full confidence. For the shopper the failure does not feel technical. It feels like the brand is unreliable, and trust is slow to rebuild.
Think about what an AI gets wrong when it guesses. An allergen left off a summary. An old formulation quoted as current. An origin claim pulled from a third-party listing that was never accurate. When an agent acts on that and recommends or buys the wrong thing, the person does not blame the model. They blame the product.
There is a balance worth naming. AI shopping agents risk being handed authority before the foundations are ready. That is the argument for a controlled resolver rather than against it. A resolver gives the machine one accurate place to look, which is what keeps the answer right. With brands already adding 2D barcodes ahead of the end-of-2027 GS1 Sunrise deadline, the barcode is going on the pack regardless. The only real decision is whether the data behind it is yours and correct.
How do you set up a GS1 resolver the right way?
Setting up a resolver starts with clean product identity. Map each product to its GTIN, express that GTIN as a GS1 Digital Link, then connect the link to a resolver that serves the right content to each audience. The work that decides your AEO value is the data behind the link, kept accurate, structured, and current.
The practical sequence is short. Get a GTIN and a 2D barcode onto the pack at your next natural print run. Express the GTIN in the GS1 Digital Link format. Connect that link to a resolver. Then structure the data behind it, including ingredients, allergens, origin, and certifications, and set the link types for each audience you need to serve.
A few mistakes cost brands the authority they are trying to build. Leaning on the free default resolver with no branded data behind it. Treating setup as a one-off instead of a living record. Letting the data go stale, which quietly teaches the machines to distrust it. And assuming the resolver is what makes the code scan at checkout, when the barcode structure already handles that. Accuracy is the whole game, a point Ian Batt learned exporting wine across regulatory regions.
How does Orijin Plus help you set up your resolver?
Orijin Plus is a GS1 Alliance Partner that manages GS1 Digital Link end to end. It generates compliant 2D barcodes, runs a managed resolver on every plan, and connects each scan to structured brand, compliance, and consumer data, so your product tells shoppers, regulators, and the machines that read it the same verified facts.
Orijin Plus generates GS1-compliant QR codes with GS1 Digital Link URIs on every plan tier. Its smart codes add a managed resolver that routes by audience, so a shopper, a regulator, and a system each get the response meant for them from one physical code. The data behind the code is built and maintained in the platform through the Traceability Hub and Transparency Hub, which is how origin, certifications, and product detail become a structured record rather than scattered documents.
Because the physical code stays stable while the data updates in real time, the record an answer engine reads is always the current one. That is the quiet advantage. The GS1 Sunrise 2027 move you are making anyway, adding a 2D barcode, becomes the same move that makes your products legible to AI. One resolver, set up once, serves compliance today and answer engines as they mature. See how one brand uses it across markets in the Small Things Wine case study, or read our guide to GS1 Digital Link.
If you are adding a 2D barcode for GS1 Sunrise 2027, decide now whether the data behind it is yours and correct. Talk to Orijin Plus about setting up a managed GS1 Digital Link resolver that serves your shoppers today and the answer engines reading your products next.
FAQ
Is a resolver required for GS1 Sunrise 2027 compliance?
No. Point-of-sale compliance is satisfied by a QR code with a properly structured GS1 Digital Link URI, which a scanner reads offline without ever contacting a resolver. The resolver is required for everything beyond checkout, including consumer experiences, compliance content, traceability, and making your product readable by AI agents.
What is a GS1 Digital Link link type?
A link type tells the resolver what a given link points to, such as product information, an ingredient list, or a certificate. The GS1 Digital Link standard defines around 60 of them. Link types are how one code returns the right answer to a shopper, a regulator, or a machine, which is why they help decide what AI sees about your product.
Will AI answer engines really use my resolver today?
This is emerging rather than fully live. As brands adopt 2D barcodes and resolvers through 2027, a GTIN-anchored resolver becomes the natural place for answer engines and agents to ground product facts. The foundation is being built now, so setting it up early is a head start, not a wasted effort.
How is a resolver different from putting schema on my website?
Website schema is page-level and can be contradicted across many URLs for the same product. A resolver is anchored to the product's GTIN and controlled by the brand, so it gives one identity-linked source of truth rather than many competing pages. Both help, but the resolver sits higher in the authority order.
Do I need to redesign my packaging to do this?
No. In most cases you add a 2D barcode at your next natural print run and connect it to a resolver. The physical code then stays the same while the data behind it evolves, so you are not reprinting every time information changes.
What data should sit behind the resolver?
The facts a shopper or a machine would want to verify: ingredients, allergens, nutritional information, country of origin, and certifications. The value comes from keeping that data accurate and current, because stale data trains answer engines to distrust the source.
Who controls the data an AI sees about my product?
With a brand-managed resolver, you do. That is the point of moving authority to the resolver rather than leaving it to whatever a model can scrape from third-party listings.
