Reuters

The post Google burning through cash with spiralling AI costs appeared first on MASSIVE News.

Your online store should make it simple for customers to find and buy the products they want. As more people use AI to help them shop, it’s important to give AI shopping agents enough clear information so they can show your products to potential buyers.

Getting your products recommended by AI isn’t about having a big brand or spending a lot on ads. AI assistants prefer stores that make things easy for them. If you provide clear, structured information, you have a better chance of reaching shoppers.

Let’s look at Answer Engine Optimization (AEO), which means setting up your store so AI tools can easily read and recommend your products.

If someone asks ChatGPT for a “cast iron skillet,” the AI creates a list of options. To do this, it checks websites for information that matches the shopper’s request. If your store is on that list, you’re reaching someone who’s ready to make a purchase.

AI models ignore your website’s design and focus on your data, but they don’t treat all content the same way. There’s a kind of “confidence hierarchy,” so where you place your product details affects whether AI agents can use them.

1. Structured data includes things like schema markup, Merchant Center feeds, and product data fields. This information is made for machines to read directly. For example, a Product schema field that says “compatible with: induction” helps AI agents match your product to someone searching for a cast-iron skillet.
2. Structured content is information that’s organized in a way that’s easy for AI to understand, like spec tables, FAQ sections, or a “Materials and care” section with bullet points. It’s much easier for agents to read a clear table than to find details hidden in a paragraph.
3. Unstructured marketing copy is the type of information AI agents trust the least, so they might ignore it. For example, if you write “Built to last generations in any kitchen,” the agent has to guess what that means and probably won’t match it to an induction-compatible skillet. Instead, use a clear spec like: “Compatible with gas, electric, and induction cooktops.”

Here’s an example showing the difference between a regular marketing description and one that’s structured for AEO.

The first version has no matchable attributes, while the second has 8. If a shopper asks for a “pre-seasoned 12-inch cast iron skillet compatible with induction,” the second version matches on 3 points, but the first doesn’t match at all.

Tips for writing product descriptions

These recommendations can really help when you’re writing product descriptions:

1. Name the use case. “Designed for high-heat stovetop searing and oven-to-table cooking” rather than “Built for the modern kitchen.”
2. Include negative qualifiers. “Not recommended for glass-top stoves” and “Not suitable for acidic foods during the seasoning period” help agents exclude your product from queries where it’s a bad fit.
3. Specify compatibility precisely, like “compatible with gas, electric, and induction cooktops.” “Not compatible with glass-top stoves” gives an agent clear data. “Works with most cooktops” makes it guess.
4. Mention who the product is for. For example, say it’s designed for home cooks switching from non-stick pans who want better heat retention for searing. More shoppers now tell AI assistants things like “I’m a beginner cook,” “I have an induction cooktop,” or “I need something oven-safe.” If your page names its audience, it gives agents additional information to match with those requests.
5. The order matters. Start with the key attributes, then share the story. AI agents read the top of your description first, while people who are interested will keep reading for more details.

AI agents look at your entire store. Here are some important areas to focus on:

1. Category pages

Many category pages only show product grids and don’t include helpful text. Adding a short paragraph at the top that answers a question like “What to look for in a cast iron skillet” gives AI a summary of what you offer.

In WooCommerce, you can add this in WP Admin under Products → Categories, and then edit the Category description. It doesn’t have to be long — just a short paragraph about use cases, key specs, and who the category is for.

2. Frequently Asked Question blocks

If you can’t fit all the product details into the main description, add FAQ blocks at the bottom of the page. For a cast-iron skillet, you could include questions like “Is this skillet compatible with induction cooktops?” or “How do I season it for the first time?”

3. Policy pages

Return policies, shipping times, and warranties are important trust signals for AI tools. Instead of hiding your terms in long legal documents, use clear labels and exact numbers, like “30-day return policy” or “ships within 2 business days.”

4. Add an llms.txt file

Make a file called llms.txt and put it in your site’s root directory at yourdomain.com/llms.txt. This plain Markdown file gives AI agents a summary of what your store sells and where to find your most important pages. llms.txt helps certain AI tools quickly understand your business by emphasizing pertinent categories and pages, making it easier for them to match your store to shopper queries. 

Currently, Anthropic and Perplexity support reading llms.txt, but Google doesn’t, so its benefits are limited to those AI platforms for now. While there is no direct evidence yet that adding llms.txt will automatically increase how often your store is recommended by AIs, it’s a simple way to organize and clarify your information for systems that use it.

Here’s an example of what the Markdown file should look like:

# Foundry Kitchen Co.
Online retailer specializing in cast-iron cookware, carbon steel pans, and cooking accessories for home cooks and professional kitchens.
## Key pages
- [Shop all cast-iron skillets](https://example.com/cast-iron-skillets/)
- [Carbon steel cookware](https://example.com/carbon-steel/)
- [Seasoning and care guides](https://example.com/care-guides/)
- [Shipping and returns](https://example.com/shipping/)
- [About us](https://example.com/about/)

Upload the file to your root directory next to robots.txt. In WooCommerce, you can do this quickly with FTP or a file manager.

If you use Yoast SEO or Rank Math, you don’t need FTP. Both added built-in llms.txt generation in 2025, starting from Yoast SEO version 22.0 and Rank Math version 1.4.0. Make sure you are running at least these versions to access the feature. 

To check your version, go to the extensions page in your WordPress dashboard. Turn on llms.txt generation in the dashboard, and it will create and update the file for you.

How to tell if it’s working

At the moment, there isn’t one perfect dashboard for tracking AI traffic, so you’ll need to do some manual checking:

1. Check your referral traffic for AI sources 

In GA4, go to Acquisition, then Traffic acquisition, and filter Session source by chat.openai.com, perplexity.ai, and gemini.google.com. You might see only a few visits at first, so don’t get discouraged. Track the numbers each month and see which product pages are getting visits.

2. Test it yourself

Every month, type a shopping question into ChatGPT or Perplexity that should show your products. Check if your store is recommended, see which competitors appear, and note the reasons the AI gives for its choices.

3. Run regular schema validation 

Go to search.google.com/test/rich-results, paste in a product page URL, and check which schema types it finds, if they’re valid, and if there are any errors or missing fields. You can also use Google Search Console and look under Enhancements in the left menu. If your schema is detected, you’ll see a Products report showing errors and warnings for your whole site.

Your next best step is to pick your highest-traffic product pages and view them through a machine’s eyes. If you find gaps where technical details are missing, rewrite them into a clear bulleted list. You can even have an LLM (large language model, such as ChatGPT) read through these pages and give its own recommendations, since they will be public.

Next, look at your highest-traffic product pages as if you were an AI. If you notice missing technical details, rewrite them as a clear list. You can also let an LLM review these pages and suggest improvements, since the pages are public.

Begin by reviewing your 10 most-visited product pages. For each, count how many matchable attributes it lists, such as materials, size, weight, use case, audience, compatibility, and any negative qualifiers. If there are fewer than five, AI agents may be skipping that page.

After you gather this information, plan to rewrite those product descriptions. Begin with your highest-traffic pages, check your AI referral sources after a month, and adjust as needed. When you make changes, focus on including any missing attributes such as size, material, compatibility, or recommended use. Look for ways to make compatibility details more specific or define who the product is meant for. This way, every update you make helps AI agents more effectively match your products to shoppers’ needs.

About

Julia Callicrate

Julia Callicrate leads Product Marketing at Woo, helping global businesses discover new ways to scale their stores with the WooCommerce platform. A storyteller at heart, Julia bridges the gap between products and customers. Prior to her work at Woo, Julia had spent over a decade as a product manager and marketer in healthcare, banking, and SaaS industries. She currently lives in Virginia with her family. Julia also has two cats that always seem to pop up on her video conferencing.

The post How to make your store readable to AI shopping agents appeared first on MASSIVE News.

AI governance is a key enterprise concern. Organizations are assembling councils, publishing principles, rolling out “approved AI tools” lists, and asking employees to opt in to acceptable use policies. In most enterprises, however, the reality is that AI is already widely embedded in employees’ daily work, often outside sanctioned channels and oversight. The visibility and control mechanisms needed to govern AI use are immature or nonexistent.

The result is a widening gap between what leadership desires for AI governance and what’s happening inside their organizations. CIOs must turn to technology guardrails capable of transporting AI governance intent from the realm of policy principles to the world of production environments, with scalable visibility and enforcement.  

Shadow AI Is the Default

Visibility is among the biggest challenges in AI governance: A recent survey found 45% of employees have used AI tools for work without informing their manager.¹ Shadow AI can take many forms, including AI-enabled web apps, browser extensions, desktop apps, and SaaS platforms. 

Shadow AI isn’t just a compliance problem — it’s a serious security and data exposure problem. Employees may paste sensitive data into chatbots, connect critical business accounts to AI-enabled workflows, grant AI applications excessive permissions, or expose proprietary corporate files to AI agents. Every AI connection creates a new identity relationship that organizations must understand and govern. A study published earlier this year found more than half of employees admit to connecting third-party AI tools with other work systems without IT department approval or oversight.²

Traditional governance and security controls weren’t built to observe and interrogate the new AI prompt and agentic interaction layer, nor were they designed to continuously evaluate the identities and permissions behind those interactions.

AI Policy Demands a Collaborative Approach

Legal and privacy teams are essential to the development of AI policy, but they can’t be the only authors. AI governance isn’t only about what’s allowed. It’s about what’s possible in the architecture, what’s safe in the threat model, and what’s useful to the business. Effective AI governance requires these stakeholders at the table:

• Business and product owners to align governance to outcomes, so controls don’t simply block innovation but shape it toward trusted, compliant, high-value use cases
• IT and security leaders to define threat scenarios (e.g., prompt injection, model supply chain risk, agent autonomy), establish controls, and ensure detection and response can extend to AI workflows
• Engineering leaders to weigh in on architectural possibilities and limitations and commit to implementing guardrails where they matter: strong identity controls, continuous authorization, logging, segmentation, safe tool use, and secure-by-default patterns in apps that call models

Determining AI governance policy is still a work in progress for many organizations. With multiple stakeholders and rapidly changing technology, it can be tricky to achieve alignment. An IBM study conducted last year found nearly two-thirds (63%) of organizations lacked AI governance policies.³ Even among organizations that reported having AI governance policies, more than half reported they lacked both approval processes for AI deployments and the technologies needed to enforce governance policy. 

The success of AI governance depends on operationalization. Few organizations today have the means to assess adherence at scale, detect violations, and continuously prove their guardrails are working. A policy that can’t be enforced becomes an artifact — useful for signaling intent but unreliable as a risk management mechanism. AI governance must become measurable: What AI tools are being used? Where is data going? Which models are connected to which business processes? Which human and non-human identities can invoke those models, access sensitive data, or delegate actions to downstream systems? What’s the rate of policy exceptions, and are those exceptions becoming the norm?

AI Agents Raise the Governance Stakes 

As AI technology rapidly changes, AI governance becomes harder. We’re moving from users asking questions of chatbots to the deployment of full-fledged AI agents that can plan, take actions, call tools, and chain tasks together.  

These agents multiply both impact and risk. They can touch more systems, execute more steps, and make more decisions faster than traditional oversight loops. Risks can go beyond bad answers to unintended actions: sending data externally, changing records, triggering financial transactions, or interacting with third parties in ways no one anticipated.

Each AI agent operates as an identity, and they rarely operate alone. They increasingly function as part of an identity chain — a sequence of humans, agents, applications, APIs, and data stores connected through delegated trust. This creates implications for identity governance. Identity can no longer be treated as a point-in-time decision. As AI agents operate continuously, inherit permissions, invoke APIs, and interact with multiple systems, identity must become a continuously evaluated security signal based on real-time context.  

The AI agent ecosystem evolves on a nearly daily basis. In the latest wave of open-source momentum, projects like OpenClaw have gained attention as developers experiment with increasingly capable agentic frameworks. Whether a given framework becomes businesses’ standard or not, the broader trend is clear: Capabilities are diffusing rapidly, and governance must account for AI tools that employees can adopt in an afternoon.

A Strategic Opening for CIOs and CISOs

Organizations that govern AI with discipline can scale it with confidence and move faster with fewer do-overs, fewer operational and security incidents, and greater credibility with customers, auditors, and regulators. CIOs, in close partnership with CISOs, are uniquely positioned to lead. Governance without security is hollow, and security without business and operational alignment fails to deliver durable outcomes.

Leaders can focus on three practical moves:

1. Enforce technical guardrails. Define what must be technically enforced (data classification rules, approved model endpoints, least-privilege access, authentication, logging, token controls, prompt and output handling) and what can be guidance. Then invest in the controls that make enforcement real.

2. Treat AI governance like an operational program. If AI governance is reviewed annually, or even quarterly, it’s already stale. Set and lead a weekly or monthly cadence with security, engineering, and business stakeholders to review adoption, incidents, exceptions, and new capabilities.

3. Define metrics and automate measurement. Governance should be provable. Track the number of AI tools in use, sanctioned vs. unsanctioned usage, sensitive data interaction rates, policy exception volume, agent deployments, and mean time to detect/respond to AI-related events. Automate collection wherever possible.

AI is moving too fast for more static, document-driven governance approaches of the past. Organizations that treat AI governance as theater will be surprised by shadow AI, agent sprawl, and incidents that were preventable. The enterprises that build guardrails grounded in visibility, identity, and continuous enforcement will earn something far more valuable than compliance: the ability to scale AI with confidence.

Additional Resources

1. Gusto, Is AI Coming for My Job? A Look Inside America’s Workplace Anxiety and What Employers Need to Know, July 14, 2025

2. BlackFog, BlackFog Research Reveals Rising Shadow AI Risks, Jan. 27, 2026

3. IBM Cost of a Data Breach Report 2025: The AI Oversight Gap

The post Why AI Governance Without Guardrails Is Theater appeared first on MASSIVE News.