The question I've been asked most this year is whether AI-written content will be penalised in search. It's the wrong question, and Google has been reasonably clear about why. Their stated position is that they reward helpful content produced for people, regardless of how it was produced. The problem was never authorship. It's that the cheapest way to produce content at volume has always produced content nobody needed.
What 'helpful' is actually measuring
Strip away the terminology and the underlying question is whether someone who lands on your page gets what they came for and stops looking. That's observable in behaviour. A page that answers the question completely, from someone with genuine experience of the topic, keeps people. A page assembled from the top ten existing results, restating the consensus in slightly different words, does not — regardless of whether a person or a model assembled it.
Why AI makes the existing problem worse
Content written by summarising what already ranks was always weak. The difference now is scale: what used to take a content farm a month takes an afternoon. So the volume of derivative material rises sharply, the bar for standing out rises with it, and the pages that survive are increasingly the ones containing something that cannot be synthesised from existing sources — original data, first-hand experience, a genuine position.
- Original data from your own product or customers — impossible to synthesise
- First-hand experience, including specifics about what went wrong
- A clear position that some readers will disagree with
- Genuine depth on a narrow question rather than shallow coverage of a broad one
The practical test before publishing
Ask whether this page contains anything that isn't already available on the first page of results. If the honest answer is no, publishing it adds nothing — not because of any penalty, but because there's no reason for anyone to prefer it. That test disqualifies a large share of what most B2B content calendars contain, which is uncomfortable and worth confronting.
What to do with the existing library
Most companies have a back catalogue of thin posts written for search volume during some earlier strategy. They dilute rather than help. Go through them: consolidate the ones covering the same ground into a single genuinely good page, substantially improve the few worth keeping, and delete the rest. Pruning a content library is deeply unsatisfying work with no visible output, and it's one of the highest-return things a B2B marketing team can do this year.
Nobody is checking whether a human typed it. They're checking whether it was worth reading — which was always the standard, just much easier to fake when faking it was expensive.
Want to apply this to your business?
Book a free 30-minute strategy call and we'll diagnose your specific growth blockers together.
Book a Free Call