Here’s how Google's helpful content update is going to make AI better

• Read original article here

Almost as soon as word broke out about Google’s latest algorithm, an update that advocates for “original, helpful content,” the questions started flying: Is this targeting AI content tools? Does this mark the end of artificial intelligence in content creation?

At Jasper, an AI Content Platform, we heard quite a few of these. Some of the questions came from our users who rely on Jasper daily to help flesh out ideas, repackage their content into different formats, and generally break through writer’s block. Others came from the broad public or pundits who are trying to figure out AI’s place in the creative world. We say to each of them and you that Google’s Helpful Content Update is a good thing. It’s a good thing for the internet broadly, and it’s a good thing for the long-term evolution of AI in content creation.

Here’s why: no one wins when the internet is littered with junk content. Like the internet itself, AI is technology that can be used to make things better or to make things worse. Humans decide which path we take. We can use both the internet and AI to create some pretty mindless things. We can also use them to remove barriers in the way of our vast creative potential. Technology is an enabler. We need to choose how to use it. In every article we write and every link we share, we need to ask ourselves: Is this piece of content adding value, or is it just filling space?

Having incentives like better distribution for higher quality content will help raise the bar for better use of generative AI. The use of AI as a tool in creative fields is still in its early days. These are the moments in which we can come together to form the standards we want to see. Standards will evolve and improve over time, and with them, we’ll see new examples of how AI can unlock creative potential. 

Dave Rogenmoser, CEO and co-founder of Jasper, put it nicely in aJasper user group the other day. Addressing questions about AI and the new Google update, he wrote, “If your AI-written content is low quality and doesn’t help readers, it’ll get dinged. If your HUMAN written content is low quality and doesn’t help readers, it’ll get dinged. How do you make sure you’re safe? Know your intended reader deeply. Write content that solves their needs and answers their questions.”

Google has long asked content creators to stop writing content for search engines and start writing it for people. But the pressure to rank in business writing is ever-present, and despite Google’s own advice, creators are still trying to crack some Konami Code of SEO in how they write and structure content. I’ve been writing professionally for decades now and have seen every variation: specific word-length targets, hyper-frequency in publishing cadence. You name it. When you take those fictional hoops and add in tight deadlines and resource shortages, you’ll likely see even the highest skilled and best-intentioned writers make bad choices.

There are already great articles about how to ground your writing in good practice for Google’s latest update. Google’s guidelines are fairly clear here. Stay away from writing about topics that fall outside of your site’s core expertise just because you think they’re traffic magnets. Don’t create bandwagon content that just repeats what has already been said. Don’t write shallow summary content that never delivers on its promise. Don’t chase arbitrary word counts. These are pretty core to good writing regardless. Now, there’s also the matter of AI-generated content. Google mentions that creators shouldn’t use “extensive automation to produce content on many topics.”  That is open to some interpretation, so let’s go a step deeper into what we think constitutes good use of AI in content creation. 

Could you write an entire article using AI commands without a single original thought? Probably. Would the resulting content be shallow and a waste of everyone’s time? Most definitely. AI tools like those within Jasper’s Content Platform are built to help people convey their ideas. But you have to have original ideas for that partnership to work. AI works best when it helps you through common barriers writers face: suggesting a transition paragraph, for example, or rephrasing a line you can’t seem to get right. Every week writers and content creators lose precious time stuck in their own work. Think about how many books have gone unwritten and how many ideas are left to wither away in a digital doc somewhere because their originators got stuck and walked away. This is where AI shines.

Google’s new update says that articles with “extensive automation” are likely unhelpful, and we agree. If you’re looking back at your content and AI has written most of it, you probably haven’t used AI well. There is no specific line denoting what an acceptable volume of AI-assisted sentences is in a given article, but this is where good human judgment comes in. Similarly, don’t take every AI-recommended line as is. If you’re blocked, pull up some suggestions, then either keep them or modify them to work. The point is not to let writer’s block stop you from communicating an idea or educating people on a topic in a way you’re uniquely qualified to do.

I used to teach a writing course at Boston University. One of the most common errors I saw in papers was when a student would cite Wikipedia or Google as the source of a particular reference rather than the original study. I still see this behavior all the time in business writing today. Content marketers will cite Statistica or another blog rather than the original source. In pulling stats, they’ll write first and add evidence later rather than starting with research. This results in cherry-picked stats that are often outdated or from a misused sample set.

 Similarly, AI content assistants – at least as they stand today – are not meant to conduct research for you. They are meant to help you package what you’ve learned from your research into a well-written article, email, or post. The good news is the time you save using AI tools to write your content can be put toward more in-depth research to strengthen the substance of the content. Research or first-hand experience is what turns a shallow, summary piece into an article of substance.

Google has struggled to shake the perception that more content is the most reliable route to more traffic. Their team has been clear in update after update that quality, intent, structure and authority matter much more than volume. And yet, companies still churn out content like it’s a race to cover every last keyword. For the “more content” brigade, AI has an obvious allure. It helps you move through content creation more efficiently, so it’s not a hard leap to assume this means you can and should exponentially grow your content output. 

But content farms are bad for consumers and writers alike. And as a result, they’re pretty damaging to the reliability of search engines and the future of AI too. There’s no doubt about it. You will get your writing done faster with AI, but we can’t lose the plot here. The benefit of becoming more efficient in your writing is not that you can suddenly churn out twice as much content. On the contrary, the benefit of becoming more efficient is having twice as much space in your day and mind to pursue and develop original ideas. If you use it right, AI should unlock creative potential, getting content creators off the content hamster wheel and onto a more strategic and editorial track. Google’s Helpful Content roll-out was completed on September 9, 2022, and while we’re still waiting to see how it will manifest in rankings, the intent here is one we support. We all want an internet with better quality content and writers who are incentivized and enabled to create their best.