The Risks of Using AI Tools for SEO Deployment and How Companies Can Better Navigate Them

As search engine optimization (SEO) has matured as a channel in the last 15 years, many businesses have gotten used to relying on organic search as a steady source of traffic and revenue. Technologies which threaten to disrupt SEO, or that promise huge SEO opportunities, are going to catch the attention of many marketers, and right now AI is both.

It’s important to remember that AI in SEO is nothing new. Google has quietly employed machine learning (ML) in its algorithms behind the scenes for at least a decade, and many SEO tools (including Moz) have provided customers with ML and natural language processing (NLP) - powered data for more than 5 years, too. What’s getting people excited in 2023 is Generative AI and the function of this kind of technology to produce text, images, audio, and video. Most famously, via OpenAI and their ChatGPT interface. This changes SEO in two ways.

How AI is changing SEO

First, Generative AI can be used as an answer engine, and as such, an alternative to web search. We’ve seen both Bing and Google experiment with leveraging this. Second, websites may start leveraging Generative AI to ideate, optimize, or even produce their content. I don’t believe that search engines producing AI-written answers are going to be as disruptive as some do, but the area many of us are focused on now is the website’s or publisher’s perspective. Specifically, how safe these tactics are.

Challenges and considerations in AI-powered SEO

The most extreme application, entirely replacing human writers and editors with AI, is not a good idea for the time being, and probably for a long time. Google’s guidelines technically allow for AI-written content as long as it is helpful but also mention the need for “experience, expertise, authoritativeness, and trust”, all of which an AI model cannot project. It’s likely not a coincidence that Google introduced a new ranking system (the “Helpful Content system”) just before the recent wave of Generative AI mass adoption and updated it very recently. It’s safe to assume that Google is better than anyone at identifying AI-written content, and indeed there are free third-party tools that are already adept at this. In addition, Google mentions devaluing entire websites if even part of them seems “unhelpful” in its announcements around the Helpful Content updates. So, unless you are in an industry where it pays to churn and burn websites and brands, this is a dangerous game to play.

How companies can navigate using AI responsibly and ethically

It doesn’t have to be all or nothing. Some brands have experimented with multi-step processes, layering AI ideation, outlines, and drafting with expert human oversight, especially at the final sign-off stage. This was the topic of one of the top talks at the MozCon conference this year, and you can watch that talk for free on our MozCon video bundle page. This can approach or even exceed the effort level (and quality) of human-written content. I suggest a sensible entry-level experiment at the ideation and outline stage. Get your keyword and market data from a traditional SEO tool (ChatGPT will claim to have this data, but unless fed via an API, it doesn’t), then give that data to the AI to suggest topics and outlines, leveraging its creativity and quick turnarounds without relying on its diligence and attention to detail.

Lastly, it’s worth taking a broader perspective for a moment. Even if you’re not worried about Google catching you out, you have to think about the user experience and ethics of questionably accurate content. Companies have to ask themselves if this is a trend they want to be part of or if they think customers will respond positively to it. Another consideration is that they should include disclaimers to inform customers of any AI-written or assisted content and think about whether that will affect how customers receive and react to the information.

In conclusion, the integration of AI in SEO strategies is becoming more prevalent among companies aiming to stay competitive in today's digital landscape. AI's ability to analyze, predict, and respond to user behavior and search patterns significantly enhances SEO effectiveness, leading to improved visibility and higher conversion rates. Companies should embrace this technology but also commit to continuous learning and adapting their strategies so that they can stay one step ahead of both the positive and negative impacts of AI on SEO.

About the Author:

Tom Capper is the Head of Search Science at Moz.