2024 has been an interesting year for the world so far. Perhaps the biggest changes, however, have not been quite as devastating as a giant hurricane, nor as dramatic as an election. No – the largest shifts have, in many ways, happened in a world not visible without the use of a device: the online world.
For anyone who has done a search on Google recently (or pretty much any other search engine for that matter), they will have seen something new and different. Yes, Google has now fully rolled out its AI summary integration at the top of the Google search with some … erm .. interesting results.
This post, however, is about more than just a new function in a search. You see, Google has been very busy as it prepares for a new coming wave of AI content. The first thing Google has done is focus less on the amount of content produced by a website or organisation and more on helpful content and the quality of that content in meeting its users’ needs.
The core issue and reason for this change is that the vast amount of AI-generated content means the amount of data that Google needs to manage is impossibly high. So, it will soon (and in some cases already) reduce the crawl budget for most websites (how many pages it will look at) and ensure that thin, unhelpful or generic content won’t be indexed. At all. I will say again, if your website has a small amount of text, produces generic content and you don’t provide helpful information – your website will not rank on Google.
Yes, like all technology we use today, the Google search algorithm has shifted to accommodate AI and ensure companies are not using it to game the system. This doesn’t mean AI-generated content isn’t capable of ranking, but that the website source and the content itself needs to play within a certain set of rules to rank effectively. Most companies are still putting out bland, business-focused content at an unheard-of rate, but little do they know their voices are being lost within an anti-AI content-generating renaissance.
This approach was fully rolled out in August, where Google added a significant boost to independent websites that promote good quality content. The content they boosted contained no ads, no sponsorships and just aimed to help people get good answers to their questions.
For many companies, especially small companies who regularly focus on providing advice, this came as a welcome change. Many saw a boost to their content and more traffic.
The only problem is that this content serves a purpose. It now feeds the AI content generation tool. Practically speaking, the summary now takes the place where organic content used to rank, pushing websites away from the top of the page – the spots that have always been most valuable for companies. You see, the way Google works has fundamentally shifted, and it isn’t going back.
This ultimately means people are less likely to visit websites and will just use the summary (from helpful websites) to get the information they need. While estimates vary, there is a significant drop in website visits and click-through rates across the board following this update. People are getting the information they need and leaving, rather than trawling through content where they are forced to endure potentially bad web practices and ads. Already it is expected that ads revenue for publishers will drop by $2 billion in the first year alone.
Of course, Google Ads aren’t quite as affected as organic traffic. Google ads can appear above the AI-generated summary and this change can, in some cases, lead to a reduced cost per click as it further separates ad traffic from organic traffic. Let’s take a breath though, and talk about AI-generated content for a moment.
It’s all the same, just different ….
Humans are reproduction machines. We take other people’s thoughts, feelings, lifestyles and we replicate them and call it our own. It’s what we do. It perhaps stands to reason then, that an AI generated by humans would tend towards a similar pattern.
One of the biggest issues faced by people who don’t often use AI, much like the challenge faced by those who have never been taken in by a confidence trick, is that it’s all new to them.
Many people who understand these common tendencies, however, can quite quickly start to see some ways of phrasing things that act as a signal that the content is nothing but a replication of other people’s work.
And so we come to the top most commonly used phrases used by LLMs. No doubt you will have seen them yourself, but regardless – you will certainly see them more often now. You might also recognise some you use yourself, but when you see so many often grouped together, that is a bit of a giveaway. .
- “In today’s fast-paced world,” – this phrase is often used to introduce discussions about contemporary challenges or advancements.
- “It’s important to consider,” – used to lead into an analysis or detailed discussion on a given topic.
- “Whether you’re a seasoned [professional/expert] or a newcomer,” – a way to address a broad audience, making content relevant to readers at different levels of expertise.
- “One of the key things to remember is,” – often introduces a crucial piece of advice or a summary point.
- “To sum up,” – commonly used to start the closing paragraph, summarising the main points of the article.
- “This guide aims to,” – used in more instructional or informational pieces to clarify the purpose of the content.
- “Let’s dive deeper into,” – an invitation to explore a topic more thoroughly within the body of the article.
Note: This summary of terms was generated (helpfully) by Chat GPT. It is quite funny to think that it already understands its own quirks and tendencies. Could this (perhaps) mean it is self-aware?!
Well let’s see what Chat GPT thinks about my article and the direction of travel here…. But I’m only giving it a paragraph, because for my money it’s far too wordy and uses waaaay too many bullet points!
[Chat GPT says]:
The article reflects on the dynamic shifts in how AI-generated content is managed and perceived, particularly by major search engines like Google. The piece cleverly satirizes the repetitive nature of AI-generated content and raises questions about the originality and authenticity of such material. It also humorously hints at the potential self-awareness of AI systems like ChatGPT, suggesting a deeper level of AI engagement with its own outputs. Ultimately, the article captures the nuanced battleground of digital content strategy, where AI’s role is both critiqued and celebrated as part of an ongoing dialogue about the future of online information authenticity and quality.
I mean, say what you like, but I think the AI likes me …