Google, OpenAI and Cloudflare Are Challenging SEO Where?

Written by Đội ngũ Marketing365, reviewed under the Content Policy of Marketing365. Last updated 15/08/2026.

Contents

SEO is entering a phase where systems choose content instead of just ranking pages
What AI Search, robots.txt, and machine traffic changes actually affect in SEO
How evidence, transparency, and the right to be selected are changing SEO standards
What SEO in Vietnam will need to change as the market shifts toward AI Search
What SEO should do to remain useful in an AI Search environment
References

SEO is shifting roles very quickly: from optimizing to rank higher to optimizing to be trusted, cited, and selected for display by systems. For Vietnamese marketers, that means SEO no longer stands alone; it has to go together with clean data, verifiable content, and clear compliance processes.

If businesses still see SEO as a position race, they will easily miss the more important part: who controls the source, who is allowed to collect data, and who is qualified enough to appear in an increasingly proactive AI Search environment.

Key points

Google is bringing Gemini 3.7 Flash into AI Mode, increasing the role of filtering systems rather than just link lists.
OpenAI says robots.txt may not apply the way many people think to fetch bots, making data-crawl control more complex.
Cloudflare warns machine traffic could be far larger than human traffic, so SEO must also account for infrastructure and machine-access risks.
For Vietnam, the advantage will tilt toward teams that manage transparency, evidence, and technical processes rather than just pumping out content.

SEO is entering a phase where systems choose content instead of just ranking pages

Three moves happening at once are changing how SEO works. Google is bringing Gemini 3.7 Flash into AI Mode in Search, making search results no longer simply a list of websites but a response layer assembled and filtered by the system [1][4]. OpenAI, meanwhile, shows that robots.txt is not necessarily an absolute barrier for ChatGPT’s fetch bot, so control over what machines read on a website is no longer as clear as before [2].

At the same time, Cloudflare says machine traffic could grow to a very large scale in the next few years, to the point where machine-generated traffic could overtake traffic from real people [10]. As machines take a bigger role in collecting, synthesizing, and distributing information, SEO can no longer ask only, “What rank are we?” The better question is: which content does the system understand as trustworthy, worth citing, and worth using as a source?

This also explains why debates around watermarking in AI content matter for SEO. If the search ecosystem increasingly needs to know who created the content, how it was processed, and whether there are signs of automation, then transparency is no longer just a communications issue; it is part of being eligible for display [3].

What AI Search, robots.txt, and machine traffic changes actually affect in SEO

The updates below are not here to report news, but to show exactly which SEO tasks each change touches directly. Every item is something that can be checked, and each one requires doing something differently.

Gemini 3.7 Flash in AI Mode: content must be written for both readers and filtering systems

Google Brings Gemini 3.7 Flash To AI Mode In Search and Gemini 3.7 Flash rolling out in Google Search show that Search is increasingly using models to process questions and choose answers [1][4]. For SEO, this means content needs to be more clearly structured, less ambiguous, and easier for the system to summarize. An article cannot just have keywords; it must have arguments, context, and enough trust signals for AI to use it without distorting the meaning.

A meeting table covered with printed documents, sticky notes, and a magnifying glass in a strategy room

robots.txt and ChatGPT’s fetch bot: crawl control is no longer simple

OpenAI Says Robots.txt May Not Apply To ChatGPT’s Fetch Bot shows that robots.txt does not always guarantee a website will be respected in the way people have long expected [2]. This directly affects technical SEO and data governance: businesses need to review how they block access, how they allow access, and how they record which bots are visiting the site. If a website has sensitive content, internal documents, or pages meant only for a specific group, relying on a single robots.txt file is not safe enough.

A server corridor with a rules sign, card reader, and a slightly open rack door

Claude AI watermarking: content provenance transparency becomes an operational signal

What A Claude Watermark Can & Can’t Tell You About Authorship reminds us that watermarking is not only about authorship but also about identifying automated content [3]. For SEO, this creates a practical consequence: businesses need to know which content is written by humans, which content is AI-assisted, and which content has been verified. As search tools and the surrounding ecosystem pay more attention to provenance, marketing teams cannot leave content production processes vague.

Machine traffic is rising fast: SEO must also factor in infrastructure and machine-access risk

Cloudflare warns that non-human traffic could rise sharply over the next few years [10]. For SEO, this is not just a reporting-traffic issue. It affects server logs, crawl-budget analysis, the risk of being over-crawled, and even security risks if unwanted bots access sensitive resources. Put simply, SEO now needs to work more closely with technical and data-security teams instead of sitting only with content.

A van, security fencing, and technicians in front of a data center at sunset

How evidence, transparency, and the right to be selected are changing SEO standards

The common thread across these developments is that they are all pushing SEO away from the mindset of “enough keywords is enough.” As search systems, crawling bots, and AI models participate more heavily, what matters is no longer the number of posts but the trustworthiness of the input. Sources such as Search Engine Journal, Search Engine Land, and Cloudflare are all showing an environment where data, access, and the way systems understand content matter more than before [1][4][10].

The right to be selected: SEO must compete on how clear the evidence is

When Google’s AI Mode filters content to answer questions, businesses are not only competing on rankings but also on the ability to be selected as a source [1][4]. That forces SEO to make authorship, citations, definitions, timestamps, and content logic explicit. The easier a piece is to verify, the better its chance of being reused by the system. This is why pages with content that “says a lot but proves nothing” will gradually weaken.

Data compliance: robots.txt, watermarking, and crawl policy are colliding

OpenAI’s fetch bot blurs the line between “blocked” and “actually safe” [2]. At the same time, Claude’s watermarking pushes the question of content origin even higher [3]. These two issues meet in one place: businesses need clearer data and content policies. SEO is no longer just about optimizing individual pages, but about managing how a website is read by machines, recorded by machines, and reused by machines.

Policy files, watermark sample sheets, and a clipboard on a bright wooden table

SEO operating costs: the biggest expense may no longer be content

Cloudflare says machine traffic is rising quickly, while Google is putting AI at the center of the search experience [1][10]. Put together, the real cost of SEO will shift from content volume to control capability. Businesses will need to invest more in crawl testing, log monitoring, data-quality checks, verification of AI-assisted content, and internal policy updates. If they do not, the website may still have articles, but those articles will no longer be assets the system can trust.

What SEO in Vietnam will need to change as the market shifts toward AI Search

In Vietnam, most marketing teams are still used to measuring SEO by rankings, traffic, and final conversions. That approach is not wrong, but it will be insufficient if AI Search changes search behavior faster. When Google uses models to answer, when bots can collect data in ways that are no longer simple, and when content transparency becomes an important signal, Vietnamese SEO needs to go beyond article optimization.

A marketing team standing outside a glass building in Ho Chi Minh City with printed papers and phones

The first thing that needs to change is the process. The SEO team has to work more closely with technical, legal, and content teams to know which pages can be crawled, which cannot, which content is AI-assisted, which content must be manually reviewed, and which data is sensitive. The second thing that needs to change is how writing is done. SEO articles should have clear structure, short sentences, concise ideas, real sources, and limited vague wording. In an environment where systems choose answers on behalf of users, the easier content is to verify, the longer it can survive.

For Vietnamese businesses, this is also the time to drop the mindset that “SEO is the content team’s job.” SEO is moving closer to trust and data governance. Whoever manages those two better will have the longer-term advantage.

What SEO should do to remain useful in an AI Search environment

Review robots.txt, access logs, and crawl policies to know which bots are entering the site, which should be blocked, and which are allowed to read.
Standardize content so it is easier to verify: include sources, authors, update dates, and clear explanations.
Classify human-written content, AI-assisted content, and sensitive content to avoid ambiguity about provenance.
Work with the technical team to monitor machine traffic, optimize infrastructure, and reduce the risk of over-crawling.

References: [1] Search Engine Journal – Google Brings Gemini 3.7 Flash To AI Mode In Search: https://www.searchenginejournal.com/google-brings-gemini-3-7-flash-to-ai-mode-in-search/585923/ [2] Search Engine Journal – OpenAI Says Robots.txt May Not Apply To ChatGPT’s Fetch Bot: https://www.searchenginejournal.com/openai-says-robots-txt-may-not-apply-to-chatgpts-fetch-bot/585864/ [3] Search Engine Journal – What A Claude Watermark Can & Can’t Tell You About Authorship: https://www.searchenginejournal.com/what-a-claude-watermark-can-cant-tell-you-about-authorship/585754/ [4] Search Engine Land – Gemini 3.7 Flash rolling out in Google Search: https://searchengineland.com/gemini-3-7-flash-rolling-out-in-google-search-485058 [10] Search Engine Journal – Cloudflare: Machine Traffic Could Hit 1,000x Human Traffic In 5 Years: https://www.searchenginejournal.com/the-biggest-ai-crawler-on-my-website-was-hunting-for-credentials/585159/

See more marketing analysis and guides at https://marketing365.vn.

Follow more analysis from Marketing365 to stay updated on the latest marketing trends.

Read more articles in the same category at Digital Trends.