Google says AI content still must meet quality standards

The quality bar did not move just because the tooling did

A book filled with AI shortcuts, factual mistakes, and fabricated quotes is the perfect cautionary tale because it shows how fast sloppy automation collapses under scrutiny. Greg Jarboe uses that controversy, and Sam Sifton’s public reaction to it, to make the larger point that Google’s standards for visibility have not softened just because AI now sits closer to the search experience.

That is the part many marketers keep getting wrong. Google’s systems still reward content that is helpful, reliable, and created to benefit people rather than content designed to game rankings. The machine has changed; the editorial lesson has not. If anything, the pressure to prove substance has gotten harder because AI can produce passable copy at scale, which makes weak work easier to flood into the index.

What Google is actually saying

Google’s guidance is blunt about automation. Automatically generated content with the primary purpose of manipulating ranking in search results violates spam policies. Google also says its automated spam-fighting systems keep operating no matter how the spam is produced, including spam generated with AI.

That matters because a lot of AI talk still gets framed as if search has become a rule-free zone. It has not. Google’s people-first content guidance still says its ranking systems prioritize helpful, reliable information created to benefit users, not content created to manipulate search engine rankings. In other words, the same old question still decides whether a page deserves attention: does this help a real person, or is it just built to perform for the algorithm?

E-E-A-T is still the backbone

Google’s quality rater update in 2022 made the company’s priorities even clearer by explicitly adding E-E-A-T, with a strong emphasis on original and helpful content. Expertise, experience, authoritativeness, and trustworthiness are not decorative concepts. They are the signals Google keeps coming back to when it wants to separate material worth surfacing from content that merely exists at scale.

That is why the current AI conversation is so revealing. AI can help assemble background, organize notes, and speed up logistics, but it cannot replace the judgment that makes a piece credible. It cannot decide which claims need verification, which sources deserve skepticism, or which details are too thin to publish. Those are the jobs that create E-E-A-T in the first place.

Why the current controversy is such a useful test case

The Sam Sifton example is useful because it shows how visible the failure becomes when AI is used carelessly. Factual errors are bad enough, but fabricated quotes are fatal. They instantly expose the difference between content that was merely generated and content that was actually reported, checked, and shaped by someone accountable for it.

That is the core editorial warning in Jarboe’s piece. Speed is not authority. Output volume is not proof of usefulness. If your content looks like the same generic material everyone else can generate, it is unlikely to earn durable visibility in AI answers, whether those answers come from Google, ChatGPT, or another system. The surface may change, but originality still wins because it is harder to fake.

How Google wants AI-assisted content handled

Google’s guidance on AI-generated content leaves room for automation, but not for hiding behind it. If you are automatically generating content, Google advises you to consider adding background on how it was created in a way that makes sense for your audience. That is not a loophole. It is a nudge toward transparency.

The practical reading is simple: if AI helped produce the content, be honest about the role it played when that context matters to readers. More important, do not let the automation become the point of the piece. The audience should still be able to see the reporting, the editing, the expertise, and the final judgment. AI can assist the process, but it should not be allowed to erase authorship.

March 2024 still matters in 2026

Google’s March 2024 spam update is another reminder that this is not a new campaign invented for the latest wave of AI hype. Google said then that its work to keep the lowest-quality content out of search continues. That line still captures the company’s posture now: the technology keeps changing, but the enforcement logic remains the same.

This is why marketers should stop treating AI content as a special category exempt from old rules. Spam is spam whether it was written by a person stuffing keywords or by a model churning out thin pages at industrial speed. Google’s systems are built to catch manipulation, not to reward the cheapest possible content factory.

What changed in May 2026

The most interesting shift is not that Google relaxed its standards. It is that Google started building more explicit routes for original work inside AI Search. In May 2026, Google announced Preferred Sources, along with new labels and carousels meant to help users find original content and firsthand perspectives in AI Search. That is a strong signal about where Google thinks value lives.

At the same time, Google’s May 2026 AI-search guidance said foundational SEO best practices still matter for generative AI features. It also said success in those features depends on creating valuable, non-commodity content and maintaining a clear technical structure. So even as the presentation layer becomes more conversational and AI-driven, the underlying requirements still look a lot like classic SEO done well: make the page easy to crawl, make the content worth citing, and make the source unmistakably credible.

What this means for publishers and brands

The practical lesson is not to produce more AI content. It is to produce better content with stronger editorial scaffolding. If the best your brand can do is the same generic explainers everyone else can generate, AI search will not rescue you. It will expose how replaceable you are.

Focus on the work that creates durable visibility:

• Original reporting, not recycled summaries
• Clear authorship and accountability
• Fact-checking that catches obvious mistakes before publication
• Specific examples, first-hand detail, and real experience
• Technical structure that helps search systems understand the page
• Transparent use of automation when it genuinely supports the audience

That is the real continuity beneath the AI hype. Google’s thresholds still reward the old virtues: evidence, credibility, usefulness, and trust. The current controversy is only making those virtues easier to see.