Insights
AI Search Depends on Infrastructure. How Exposed Is Your Marketing Strategy?
Every AI answer needs computing power. Behind the interface are servers, electricity, cooling systems and decisions about who gets access to them.
For a business investing in AI visibility, that creates a planning question: how much of your customer discovery depends on platforms whose infrastructure and commercial decisions you cannot control?
AI search risk deserves attention as you decide where to put your marketing budget. Understanding your exposure starts with knowing what these platforms contribute to your business and how you would respond if that contribution changed.
The infrastructure behind the answer
In Denmark, pressure on the electricity grid has led to a new order of priorities for connection requests. The Copenhagen Post reported in October 2026 that critical public services and green transition projects would receive priority, with data centres among the large electricity users in the fourth category.
This concerns access to capacity for new connections. It gives us no basis for predicting changes to how ChatGPT or Google cite a company.
Capacity constraints are also visible in providers’ own reporting. In its April 2026 earnings call, Microsoft said demand for Azure continued to exceed available capacity. The company expected to remain capacity constrained at least through the end of 2026, despite further investment. It also described balancing supply between Azure, its own applications, research and server replacement.
That is a concrete example of how infrastructure shapes a provider’s decisions. Extending it to predictions about citations, rankings or referrals would require separate evidence.
For marketing teams, the practical issue is dependency. Your ability to reach buyers through AI sits within a service another company operates.
Where AI search risk enters your marketing plan
You can invest in better content, make your website easier to access and publish evidence of your expertise. The platform still determines how its answers appear and how users reach your business.
When planning, consider scenarios such as:
- Changes to the prominence or format of source links.
- Changes to subscription prices, usage limits or regional availability.
- New interfaces that alter how buyers compare suppliers.
- Different ways of presenting recommendations, commercial information or advertising.
These are scenarios to consider. Their likelihood and impact will vary by platform.
The useful question is how your business would respond. If a platform sent fewer visitors next quarter, would you know? Would it affect qualified enquiries? Could other channels sustain your pipeline?
A company can have limited AI referral traffic and still influence buyers through information they encounter in an answer. It can also receive plenty of visits that lead to little business. Visibility, traffic and commercial value each need their own measurement.
Four ways to understand and reduce your exposure
1. Keep your website useful and accessible
Your website gives buyers a place to check your services, experience, pricing and suitability. It also provides material that search systems may use when answering questions.
Start with the pages that matter commercially. Can people and relevant crawlers access them? Is the important information available as readable text? Can visitors move easily from an explanation to a case study, service page or enquiry?
For Google’s AI Overviews and AI Mode, Google’s guidance points to established SEO practices. A supporting page must be indexed and eligible to appear in Search with a snippet. There are no additional technical requirements or special AI files needed for inclusion.
Other platforms have their own access controls and crawler behaviour. Review those individually when deciding which systems you want to access your content.
Put your effort into clear pages, reliable information and a site that works for the people considering your business.
2. Publish evidence from your own work
A broad claim about expertise gives a buyer little to evaluate. A documented example gives them something concrete.
Explain how you solved a problem, what the constraints were and what changed. If you publish results, include the period measured and enough context to understand the numbers.
Useful material might include:
- A case study with a defined starting point and outcome.
- An explanation of how you price a service.
- Original research with a clear method.
- Answers to questions clients ask during a buying decision.
- A practical account of your process, including its limitations.
This material gives you a stronger foundation across your website, sales conversations, proposals and other communications. It also gives anyone describing your business something specific to work with.
Publish information you can stand behind and keep it current.
3. Measure visits, mentions and enquiries separately
Start with referral visits from identifiable AI sources in your analytics. Then look at what those visitors do: which pages they read, whether they enquire and whether those enquiries become qualified opportunities.
Treat referral data as a partial view. A buyer may discover you in an AI answer and return later through a branded search or a direct visit. Some visits may arrive without a usable referral source.
Google’s AI search features need separate care. Google includes traffic from AI Overviews and AI Mode within its overall Search Console reporting for web search. That limits how precisely you can isolate their contribution.
Combine several sources of information:
- Analytics for identifiable visits and conversions.
- An optional “How did you hear about us?” field.
- Notes from sales conversations about how buyers researched suppliers.
- Repeated checks of relevant AI answers.
For those checks, keep a record of the prompt, platform, date and available settings. Use questions a potential customer might actually ask. Record mentions, descriptions, citations and factual errors.
A fixed set of prompts gives you a sample to compare over time. Results can vary between runs and users, so interpret changes with care.
Together, these measures help you understand both the attention you receive and the business it supports.
4. Develop several routes into your business
An email audience built with consent, repeat customers, referrals, partnerships and events can give buyers different ways to reach you.
Each involves dependencies and maintenance. What matters is having a mix that can support your business when one route becomes less effective.
Be precise about control. Your website and contact database give you substantial control over your information and relationships. A LinkedIn presence remains subject to LinkedIn’s distribution and product decisions. It can still be valuable within that mix.
Choose channels your team can sustain and your customers use. Assess them by the quality of the relationships and opportunities they create.
A practical AI search risk check
Review these five questions with whoever manages marketing and sales:
- What evidence do we have that buyers discover or evaluate us through AI?
- Can we connect identifiable AI referrals to qualified enquiries and revenue?
- Would we notice a significant fall in those referrals within a month?
- Can buyers verify our most important claims on our own website?
- Which other channels could support our pipeline if AI discovery became less effective?
Uncertain answers reveal gaps in measurement or planning. They give you a starting point for investigation.
Where you can estimate AI’s contribution to qualified opportunities, model a reduction and discuss the consequences. Use your own business data to decide how much attention the risk deserves.
AI visibility can earn a place in your marketing strategy. Give it a clear role, a measurement plan and a level of investment you can explain. Keep building the evidence and relationships that help customers choose your business.
Frequently Asked Questions
What is AI search risk?
AI search risk is the exposure a business takes on when part of its customer discovery depends on AI platforms it does not operate. Changes to how those platforms show sources, price access or present suppliers can affect how buyers find you, regardless of the quality of your content.
Will infrastructure constraints change how AI platforms cite my company?
There is no evidence of that so far. Capacity limits affect how providers expand and allocate their services. Any specific effect on citations, rankings or referrals would need to be shown separately before it should change your strategy.
Do I need special files or markup to appear in Google’s AI Overviews?
No. According to Google, a page needs to be indexed and eligible to appear in Search with a snippet. There are no additional technical requirements for AI Overviews or AI Mode. Established SEO practices still apply.
Can I see AI Overviews traffic separately in Search Console?
Not as a separate figure. Google includes traffic from AI Overviews and AI Mode within the overall web search data in Search Console’s Performance report. Combine it with analytics, enquiry data and sales notes to build a fuller picture.
Should we reduce our investment in AI visibility?
Not necessarily. Base the decision on your own data: what AI contributes to qualified opportunities and how your pipeline would cope if that contribution fell. The aim is a clear role and a level of investment you can explain, alongside other channels.
Sources
- The Copenhagen Post: Energinet to prioritize addressing public service queries as power grid capacity runs short (21 October 2026)
- Microsoft: FY26 Third Quarter Earnings Conference Call transcript (29 April 2026)
- Google Search Central: AI features and your website
Put it into practice
Know what matters now. Fix what matters first.
No spam. No commitment. Just a straight answer.