Artificial Intelligence and Search: Our Key Takeaways From Google’s Recent Statement

You may have heard recently that the internet has been flooded with news all around AI (Artificial Intelligence). Whether you’re a PR expert, a content creator, developer or SEO, it will be hard to escape the sheer amount of content pumped out into the digital sphere.

It’s surreal when a topic usually geared towards Digital Marketeers makes waves in other industry pools. It’s even more surreal when it starts to crop up in civilian circles and news outlets that quite frankly, you didn’t even realise were paying attention (here’s looking at you iflscience and Ladbible).

AI has always been a hotly debated topic amongst SEOs and content creators, with many worrying that eventually it could replace what we do. Personally, I don’t see that happening anytime soon, or at all in fact.

It comes as no surprise that the bigger the wave created, the bigger (and potentially more destructive) the break and the crash. We only need to take a look at what happened with Alphabet a couple of days ago and their $100 billion loss in market value due to the new chatbot sharing inaccurate information in a promo video (Source: Reuters). I think it’s safe to say, there’s still a long way to go!

With the most recent statement given by Google about AI and Search, I think there are definitely some key takeaways that people need to understand when it comes to their interplay:

  • Do as Google have always said and prioritise high-quality content (Regardless of whether humans or machines generated it)
  • Follow and Demonstrate E-E-A-T (Expertise, Experience, Authoritativeness and Trustworthiness)
  • Using automation or AI to manipulate ranking is considered a violation of Google’s Spam Policies

Some additional thoughts we’ve had internally:

  • There is a key distinction between spammy AI generated content and useful content – Not all AI generated content is actually spam, As their statement suggests, “Google realizes not all use of automation and AI-generated content is spam. For example, publishers automate helpful content such as sports scores, weather forecasts, and transcripts” (Source: Search Engine Journal).
  • The “Who, How and Why” update to the “Creating Helpful, Reliable, People-First Content Page (Source: Google)
    • “Who” refers to the person who created the content – a byline or background info of the author should be mentioned, and there should be an author page available giving a background in their areas of expertise. Authorship information should be clear and accessible.
    • “How” – How was the content created? Was a test carried out, and if so clarity needs to be present around results and photograph evidence etc. In this instance with AI, Google wants people to share details of the process on how the content is written. This is a statement straight from the horse’s mouth: “Sharing details about the processes involved can help readers and visitors better understand any unique and useful role automation may have served”. It’s important to remember “AI or automation disclosures are useful for content” – basically, if you’re using AI to write content, you need to let readers know.
    • “Why” – What’s the purpose of this content? Is it to help people? Inform people? Regardless of whether AI created the content or not, what purpose does this serve? Here is another all important quote: “If the “why” is that you’re primarily making content to attract search engine visits, that’s not aligned with what our systems seek to reward”.

Considering all of the above, I’ve got a few key takeaways to put forward:

  • Google as a Search Engine is getting smarter.
  • AI content, although useful, is useful only in certain scenarios and at this point, I don’t believe will take over an SEO’s or content writers’ job – there’s still far too much that can go wrong (case and point, the drop in Google share price explained above).
  • We predict that in the future Google will make publishers and content authors tag up content that has been created through automated/AI means. Like we have to do with paid for links / affiliates.
  • The helpful content system which was introduced last year stated clearly that websites need to focus on people-first content. Although AI could replicate this to a certain extent, it cannot replicate human creativity and expertise to produce high-quality, informative content. AI could produce content which might be helpful but might lack the expertise touch.

This isn’t to say that AI content wont be useful in the future – How AI will be utilised is going to be key. Lots of testing and learning will need to be carried out to ensure that the content created is helpful, but also safe, where it needs to be. Don’t forget Google AI’s third principle – “Be built and tested for safety” (Source: Google AI Principles).

Google have their mission statement when it comes to Search: “to organize the world’s information and make it universally accessible and useful” (Source: Google) and with the release of Bard, and their lightweight language model of laMDA, Google are starting to test and learn themselves, in order to improve on quality and speed.

Worried about AI generated content and what it could mean for you and your industry? It’s a scary concept, and we here at The PHA Group have a highly-skilled and knowledgeable SEO team ready to help. Get in contact with us today and see how PHA can help your business.

Get in touch with the team