You've heard AI bots can't read your JavaScript. Profound now sells the fix. On October 11, 2026, it launched Dynamic Bot Rendering, which gives AI bots their own fully rendered copy of a page while people get the normal one. Gemini is one of the bots a brand can pick.

Profound also says its next step is to serve different versions to different bots. That is where Google comes in. Google says Google-Extended, the control for Gemini, has no user agent of its own.

My read is that a page built for Gemini lands in front of Googlebot, where Google's cloaking rules apply. Rendering the same content for bots is fine. Serving Google a different version is where I would stop.

The launch post lays out the product and what Profound plans to ship next.

The post says most AI bots fetch the initial HTML response and have no browser engine to run a script. Pricing pages with tabs that render on click, or pages where third-party scripts write in content after load, can look empty to them. Profound adds that Cloudflare reported bot traffic overtaking human traffic in June 2026.

The fix has two parts. Bot View, inside Profound Pages, shows what bots see next to what a human sees. Dynamic Bot Rendering then creates what the post calls "a separate, fully-loaded copy of the page" and hands it to bots. A brand picks the scope, from the whole site down to one page, and picks the bots: ChatGPT, Perplexity, Claude, Gemini and more. It is live for customers on CloudFront, Cloudflare or Vercel.

Disclosure: my agency, ViewEngine, sells AI visibility work and builds its own tracking tools. That puts it in the same market as Profound.

Profound writes: "With Dynamic Bot Rendering, AI and human visitors can see two different versions of the same page, one built for people and one for bots, without a single change in your CMS."

Then it names the next step. Profound writes: "Next, your AI Marketer will be able to optimize content, A/B test, and even serve different versions to different bots, without affecting the human experience." The launch post names no guardrail for that plan.

Where Google's rules start

Google has a page on this pattern. It calls dynamic rendering "a workaround and not a recommended solution."

The same documentation draws the line. Google Search Central writes: "Using dynamic rendering to serve completely different content to users and crawlers can be considered cloaking." A bot copy that matches the human page is rendering. Different substance can be cloaking.

Now the Gemini checkbox. Google's crawler documentation says: "Google-Extended doesn't have a separate HTTP request user agent string. Crawling is done with existing Google user agent strings; the robots.txt user-agent token is used in a control capacity."

Profound's post does not say which user agent its Gemini option matches. If it matches Googlebot, Google's spam policy applies. That policy ties cloaking to different content for users and search engines "with the intent to manipulate search rankings and mislead users." Parity shows no such intent. A Googlebot-only version written to win citations is a harder case to defend.

The case for doing it anyway

Some practitioners think the rules are on their side. Mike King, founder of iPullRank, made the case in a May 14, 2026 post that AI crawlers are not bound by Google's guidelines. He wrote: "To recap: the AI crawler ecosystem doesn't render JavaScript, doesn't follow Google's guidelines (because they aren't Google), and is actively used by your competitors to harvest and rewrite your work."

I think he is right about whether Google's rules reach ChatGPT, Claude and Perplexity. His post says nothing on whether a different version earns more recommendations, and I haven't seen evidence that it does.

Glenn Gabe, an SEO consultant, summarized a June 22, 2026 remark from Google's Brendon Kraham this way: "Don't optimize for bots. Optimize for people."

Where I draw the line

Bot rendering earns its place when a page is an empty shell until a script runs. Giving bots the rendered version of that same page closes an access gap, and Google's own documentation treats similar content as acceptable.

A different version aimed at Google's crawler is another matter, and Gemini reads through that crawler. I would keep Gemini out of any bot-specific rule until Profound says what it sends Googlebot. A copy of your page that only Google's crawler sees is the one thing in this launch I would not want to explain to Google.

Free newsletter

The Sherpa Brief

One email a week on what changed in AI search, what it means, and what to do about it. Written for marketers, SEOs, and publishers.

Free. Unsubscribe in one click.