On today’s internet, almost everything is beholden to algorithms, which in turn respond to users’ behavior. AI influences are also now normal in social media feeds. 
 
Most people feel confident they can spot AI, but studies are revealing people are actually much worse at that than they think. 
 
With this in mind, more and more, you’re competing as much against AI “slop” of low quality, mass-generated content that’s surface level at best and straight-up factually incorrect — sometimes dangerously so — at worst. 
 
So, in a sea of noise and information pollution, how does that change your own approach to your content, how to stand out, and how to strategize to that end? Let’s check it out! 

The stats: How much content is AI now, and what are people noticing (and missing)? 

The proliferation of AI has meant that an increasingly larger share of social media content is AI-generated, unfortunately in the process obscuring organic content and information put out by actual people. This means: 

• 15% of Reddit posts 
• 54% of LinkedIn posts 
• 360 million+ Facebook posts 
• 330 million+ Instagram posts 
• 5% to 15% of new YouTube uploads 

…are generated by AI — and this is only counting that which is labeled as such via these platforms’ specification requirements. It doesn’t count total unlabeled AI, the scope of which is still unknown. 

Basically: A huge amount of social media is AI-generated — enough where this new reality is something we all have to adapt to as both individual users of the internet and as companies trying to reach people who would find our products or services helpful. 

An extra layer of problems 

This by itself has created a new ecosystem that’s increasingly difficult to navigate, but there’s an additional layer that complicates things even more: People are very confident in their ability to spot AI, while actually being quite poor at it. 

For example, a Microsoft study found that 89% of people reported they were confident they could spot AI or deepfake videos — however, in actuality, only 42% of those respondents were able to identify AI images, and only 25% were able to identify AI in high-quality video. 

This is obviously extremely worrisome for many reasons outside our scope, but one of the things it’s complicated in this vein is exactly within our realm of expertise, and that’s what we’ll talk about further: That is, digital marketing by producing quality, useful content to expand and engage your user base. 

First, let’s look at the mechanics behind what’s happening now. 

How organic content is served now 

In order to best adapt to this new reality, we have to understand how things work. 

Currently, this starts with the inventory and eligibility of your content; when you post something, the platform adds it to a pool of eligible content to surface to users. Each platform “ranks” its content differently — AKA, in effect, the algorithm — where the better your content ranks, the more likely it is to be served to someone. 

On the user side, a platform’s internal algorithm predicts what each user is most likely to “value,” and then accordingly conducts a personalized ranking of content to that end. 

Of course, the issue with this — or advantage, depending on if you’re able to harness it successfully — is it creates feedback loops that funnel individual users into increasingly granular, specific rabbit holes that can be difficult for them to break out of, or you the content maker to break into. When people discuss online “echo chambers,” for example, it is essentially this process that is the culprit. 

Another way of conceptualizing this is via five phases of content serving: 

1. Signals gathering: The platform gathers signals to infer a user’s preferences, such as hook strength (measured through various means), watch time, swipe-aways, saves and shares, comments, and other forms of engagement. 
2. Micro testing: Using the interactions’ strength of a smaller follower sample, the platform infers how your content is performing and, if it deems it to be above average, will push that content wider. 
3. Content matching: Using data points from your contents’ interactions and others, the platform clarifies likely topics and overlap, as well as how that relates to other viewer behavior, which niches and clusters it overlaps with, and subsequently to whom it is relevant. 
4. Push phase: The platform pushes your content more towards non-followers based on implied interests and related in-groups. 
5. Decay phase: Reach plateaus and slows, with new content prioritized. 

What all this means for brands 

Because of how this ecosystem of organic content functions plus the effects of so much AI-generated content essentially flooding it at the same time, it means you’re effectively competing with more content than ever, all for the same amount of eyes, most of whom are inadvertently in some level of feedback loop that’s serving them primarily, well, not your content. 

Put more simply: You are now competing with more of the internet, for more of the internet. The ceiling is much higher, but the floor is also much lower. 

In practice, it means that creativity — including new perspectives, expertise, and generally just thinking outside the box — and retention beat just cadence and churning out volume. Alternatively: You can never “out-volume” the AI machine whose purpose, boiled down to it, is creating volume, but it’s the quality and creativity that you will always be able to beat it at. 

To that end, your focus within this ecosystem must shift to high-intent signals within user behavior that imply genuine interest. For example, watch time, shares, saves, and meaningful comments, among others. 

At the same time, you need a budget to support these efforts with a little boost. 

Three practices for modern social strategy 

So, what would we suggest for how to concretely work towards that? 
 
Here are three best practices for the foundation of your work: 

1. Maintain an editorial calendar and establish a reliable cadence for posting. Whether that’s once a week, three times a week, etc. — it’s better to be consistent than to fluctuate, even if the latter would produce greater overall volume. At Fujisan, we like to make a calendar one month at a time, so that if there is anything topical that comes up, you can pivot or at least address it in some way. 
2. Build a core “story,” then adapt it to each platform. Put another way: Understand and clearly establish your voice, but know that your content for, say, YouTube will utilize that voice differently than on your blog. Some crossover content between platforms is totally okay — sometimes even ideal! — but just copy-pasting your content to different platforms isn’t going to work out as well because users interact differently with different platforms, just as different platforms interact differently with different content. 
3. Engage with users. Or, don’t “post and ghost.” Check your notifications and interact with those who are interacting with you! Not only is it good “customer relations” on a human level, but it’s good for platforms’ measurement and ranking systems too. 

All together 

To kickstart your efforts, audit what of your current content is earning the best retention and engaging people in such a way that shows they have high intent towards your work — metrics like high watch time, shares, saves, meaningful (not spammy!) comments, and even messages. 

Further, choose one or two primary platforms to really focus on. Make this decision based on audience behavior on the different platforms, and don’t just do “copy-paste” posting between your platforms. 

Build an editorial calendar to help organize your schedule, and ideally this can also include repeatable content series. Engage consistently with your readers and viewers from there, and then use paid “boosts” selectively for posts, videos, and other content that’s already shown itself to be a winner organically. If you ever need help with your organic social strategy, our team at Fujisan is here to help!