Google’s Standards Stay, But AI Makes Them Harder to Ignore

When Sam Sifton, the host of The New York Times’ The Morning newsletter, sent out a recent note to subscribers, the subject line was deliberately provocative: “Who’s Writing This?” The question arose from a new book, The Future of Truth by Steven Rosenbaum, which was produced with significant AI assistance. A Times review uncovered more than half a dozen fabricated or misattributed quotes that the AI had hallucinated, including one falsely pinned on tech journalist Kara Swisher. Swisher’s reaction was blunt: not only was the quote wrong, she said, but “I also sound like I have a stick up my butt.”

Rosenbaum’s justification that these hallucinations “serve as a warning about the risks of AI-assisted research and verification” would carry more weight if it appeared in a book that wasn’t itself a cautionary tale.

Sifton used the moment to deliver a direct message to his readers: The Morning is written by humans, for humans. His team might use AI to locate information that is then verified elsewhere, or to streamline editorial logistics and free up time for deeper reporting. But the core tasks of forming ideas, asking questions, performing deep reading, and crafting the final prose remain squarely in the hands of journalists. “I write fueled by adrenaline and fear of errors,” Sifton told his audience. “And I promise you that will never change.”

What Google Has Actually Said About AI Content

Back in February 2023, Google’s Danny Sullivan and Chris Nelson published the company’s official stance on AI-generated content. That position has remained largely unchanged, reinforced most recently by Matt Southern’s coverage of Google’s new AI search guide. In short: Google’s ranking systems are designed to reward original, high-quality content that demonstrates E-E-A-T (expertise, experience, authoritativeness, and trustworthiness). The focus is on the quality of the content, not the method of its production.

At first glance, this sounds like an open door for AI-generated material. But that impression is misleading. The guidance explicitly states that using automation to create content with the primary goal of manipulating search rankings violates Google’s spam policies. The search giant draws a useful analogy: about a decade before the 2023 guidance, there were widespread concerns about content farms that churned out massive volumes of human-written articles. No one suggested banning all human content. Instead, Google improved its systems to reward quality. The helpful content system, the E-E-A-T framework, the information gain patent, and the ongoing updates to the Quality Rater Guidelines through 2025 are all part of the same enforcement mechanism, now operating at a higher level of sophistication.

Rosenbaum’s book is precisely the kind of content Google’s systems are built to identify and discount. Not because it used AI, but because it used it carelessly, without the verification, original reporting, and editorial accountability that Google’s quality signals are trained to detect.

Sifton’s newsletter, by contrast, is exactly the kind of content those same systems are built to reward. Not because it is human-generated, but because it is produced by people with genuine expertise, direct experience, and accountability to a specific audience. It is built by humans, for humans, in the very sense Google’s helpful content guidance has always intended.

Will Sifton’s Letter Change Anything?

The central question here is whether Sifton’s examination of AI’s expanding role will alter what Google does, how practitioners write for AI, or how they succeed in AI visibility. The honest answer is no, not directly, and that is precisely the point.

Google’s guidance has been consistent since February 2023. It was consistent before that in spirit, going back to Panda in 2011, through E-A-T, the Helpful Content Update in 2022, and the transition to E-E-A-T later that year. What changes is only how sharply people perceive it on the horizon.

What Sifton’s letter accomplishes, which Google’s technical documentation cannot, is to make the human cost of the alternative legible. Rosenbaum’s hallucinated quote from Kara Swisher is not an edge case or a technical glitch. It is what happens when thought-making is entirely outsourced, when questioning stops, when no one is writing fueled by adrenaline and fear of errors. It is a book about the future of truth that cannot be trusted.

For SEO professionals, the practical implication has not changed since Amit Singhal’s 23 Panda questions in 2011. Does the article provide original content or information, original reporting, original research, or original analysis? Does it have the kind of quality you would expect to see referenced by a magazine, encyclopedia, or book? Would you be comfortable handing this to your editor and putting your name on it?

Sifton’s promise to his readers is that he would. That accountability is not a stylistic choice. It is the entire mechanism by which trust is built with an audience, and by which Google’s systems learn to surface content worth surfacing.

The Real Takeaway

AI is not indifferent. It is responsive, adaptive, and improving faster than any previous technology transition in the industry’s history. That is exactly what makes it useful, and exactly what makes the question of how you use it so consequential.

But the standards that determine whether content earns trust, from readers and from Google’s ranking systems alike, do not move on AI’s schedule. They have been moving in the same direction for as long as Google has existed. Every approach that has assumed those standards would yield to scale, to automation, or to the next optimization trick has found the same result.

They don’t yield. They move right along as though nothing happened.