Something changed on social media this year, and most business owners have no idea it happened.
You’ve probably heard the noise about AI generated content. Deepfakes, fake influencers, AI images that look a bit too perfect. It’s easy to switch off when the conversation gets that dramatic. But underneath the noise, something much more practical shifted, and it directly affects any business posting on Instagram, Facebook, TikTok or LinkedIn right now.
Platforms are actively scanning your content to work out if it was made with AI. Not asking. Scanning.
Here’s what’s going on, in plain English, and what to actually do about it.
What’s actually happening
For years, if you used AI to help write a caption or generate an image, nobody really knew unless you told them. That’s no longer the case.
Most photos and videos now carry an invisible digital “receipt” attached at the moment they’re created, called a Content Credential. It records whether the image came from a camera or was generated or edited by AI. Newer phones are starting to attach this automatically, and every major AI tool does the same. Platforms read that receipt the moment you upload.
On top of that, from 2 August 2026, a new EU law requires AI generated content to be clearly disclosed. A similar law covering AI generated performers came into effect in New York in June. TikTok, Meta, YouTube and LinkedIn have all rolled out their own labelling systems as well, and they apply everywhere, not just to businesses in those regions.
In short: the assumption that AI use is invisible unless you say so is no longer true.
Why this actually matters for your business
This isn’t really about rule-breaking or getting banned. Most business content isn’t going to trigger a penalty. The bigger issue is quieter than that.
Platforms don’t just label AI content, some of them adjust how far it travels. Meta, for example, can reduce the reach of content it judges to be AI generated and misleading, sometimes significantly, without ever removing the post or telling you why your numbers dropped. Your content stays up, your account looks fine, and it just quietly performs worse.
There’s also the trust side of it. Customers are getting sharper at spotting content that feels hollow, generic or overly polished in that unmistakably AI way. Once someone clocks that a business’s “personal” story or “authentic” photo wasn’t real, that trust is hard to win back.
What this means if you’re an Australian business
Here’s the bit most Aussie business owners haven’t heard yet: we don’t have a law forcing AI disclosure the way the EU and New York now do. So it’s tempting to think this doesn’t apply to us. It does, just for different reasons.
Two things to know:
The platforms don’t care where you’re based. Meta, TikTok, YouTube and LinkedIn apply the same AI detection and labelling rules worldwide. An Australian business posting from Brisbane gets scanned exactly the same as one posting from Berlin. There’s no local exemption.
The government has already told us what “good” looks like. In late 2025, the National AI Centre (part of the Department of Industry, Science and Resources) released official guidance for Australian businesses called “Being clear about AI-generated content.” It’s not a law with fines attached, but it lays out a sensible, common sense approach: the more realistic the AI content and the more it could influence someone, the clearer your disclosure should be. A quick AI-assisted caption tidy up needs basically nothing. A fully AI generated “customer” or a realistic AI image of your product in a setting it’s never actually been in needs to be clearly flagged. [CONFIRM]
There’s also a simpler backstop that already applies regardless of any AI-specific rule: Australian Consumer Law. If AI generated content misleads someone about your business, your product or a testimonial, that’s already covered under existing misleading conduct rules, AI or not. This part isn’t new, it’s just more likely to get tested now that AI content is everywhere.
The practical takeaway for a local business: you don’t need to panic about a looming Australian law, because there isn’t one yet. But “no law” doesn’t mean “no consequence.” The platforms are already enforcing their own rules on you, and honesty about AI use was already good practice under our existing consumer law. Getting ahead of it now costs nothing and puts you in good stead either way.
The mistake we’re seeing businesses make
It’s not “using AI.” AI is genuinely useful, and almost everyone uses it somewhere in their workflow now, even if it’s just for a first draft or an idea.
The mistake is using it to skip the thinking, not just the typing. Letting AI write the whole caption with no edit. Using a stock AI image because it’s faster than a real photo. Posting something with no human judgement over whether it actually sounds like the business, or whether it’s even true.
That’s the version that erodes trust, and now, it’s also the version platforms are getting better at spotting.
How to use AI in a way that actually helps your business
The businesses getting real value out of AI right now aren’t the ones avoiding it, and they’re not the ones handing everything over to it either. They’re the ones using it as a tool with a person still firmly in charge of the outcome.
A few practical shifts worth making:
- Use AI for speed, not for voice. Let it help you get from blank page to first draft faster. Then rewrite it in your own words, your own tone, the way you’d actually say it out loud.
- Keep real photos and real people in the mix. A genuine photo of your team, your space or your work still carries more trust than a generated one, and it always will.
- Be upfront when it counts. If you’re using AI generated visuals in a way that could be mistaken for real, a quick, simple disclosure costs you nothing and builds trust rather than costing it.
- Have a human make the final call. Before anything goes out, someone who actually knows the business should be reading it and asking, “does this sound like us, and is it true?”
This is really the whole idea behind how we work. AI can absolutely make execution faster. It should never be the one deciding what your business sounds like, or whether something is worth saying at all. That’s still a human job, and it’s the part that actually builds trust.
Frequently Asked Questions
Do I have to label every post that used AI in any way?
No. The rules are mainly aimed at content that’s realistic enough to be mistaken for something real, like AI generated people, voices or scenes. Using AI to help brainstorm a caption or tidy up wording isn’t what these rules are targeting.
Will using AI get my account banned?
Unlikely for most everyday business content. The more common outcome is reduced reach on flagged posts, not removal or bans, which is exactly why it’s easy to miss.
How do I know if an image I’m using has AI metadata attached?
There are free tools that let you check, such as the Content Authenticity Initiative’s verification tool. Worth a quick check if you’re using AI generated images regularly.
Is there an actual Australian law requiring AI disclosure?
Not yet, no fines or legislation specifically for this. There’s official government guidance from the National AI Centre recommending disclosure, and existing Australian Consumer Law already covers AI content that misleads. On top of that, the platforms enforce their own global rules regardless of where you’re based, so it’s worth treating good disclosure as standard practice either way.
Does this mean we should stop using AI altogether?
No, and we wouldn’t recommend that either. Used well, it’s a genuine advantage. The businesses who’ll do best from here are the ones who know how to use it properly, not the ones who avoid it.
