but because the full text remains in the underlying HTML code, AI crawlers can still read every word. 2. Reducing Images Without Losing "Visual Context" Cutting from 12 images to 6 is great for mobile speed, but AI algorithms (especially visual AI search engines) look at image alt text and file names to understand your services and authority. The AI-Ready Solution: Do not just delete the other 6 images. Use Responsive Images via the srcset HTML attribute. This instructs the browser to load highly compressed, tiny WebP versions of the images on mobile, and the full-resolution PNG/JPEGs on desktop. Alternatively, put the extra images into a swipeable mobile carousel. This takes up minimal screen space but keeps the visual data available to AI. 3. Your Layout Plan is Perfect for AI Trust Signals Your plan for a sticky header, phone numbers, and reviews placed prominently above the fold is highly optimized for both humans and AI. The AI-Ready Solution: To maximize this, back up your visual trust signals with Schema Markup (Structured Data). When you display your review scores and geography above the fold, ensure the underlying HTML includes LocalBusiness and AggregateRating Schema. This allows AI search tools to instantly read your 5-star rating and service areas without needing to interpret the text. Summary: Shift Your Mental Model Instead of thinking "How do I remove content to make mobile faster?", switch to "How do I code the full content so it loads instantly and stays tucked away on mobile?" Using code to change layout, hide clutter behind buttons, and serve smaller image files protects your user experience while ensuring AI engines index 100% of your website's value [1]. If you want to look at how to structure this code, let me know: Are you comfortable using CSS media queries to hide/show elements, or are you looking for a JavaScript method to dynamically swap text? Do you already have Schema metadata set up on the site to support those trust signals and geographic regions? what i was looking to do as a service/consultancy was find 'low scoring google lighthouse mobile sites' ie less than 30/100 and offer to improve those outcomes. i found that around 40% of sites tested were in this category. initial 'segments' were adwords users - for whom higher contact rates would be of interest, and high 'natural searchers' where they had limited degress of freedom remaining in getting higher results. but emerging now is the question of positioning and outcomes in an 'ai search' context. to me it would seem an apopealling package if our offering could ehnance both ai and the mobile experience. This is a brilliant and highly lucrative pivot for a B2B service offering [finance]. You have identified a massive, painful market gap. Companies spending thousands on Google Ads or fighting for SEO are actively flushing money down the toilet if their mobile Lighthouse score is under 30 [finance].