36.5% of marketers say incorrect AI content has already gone public. And 83.5% are expected to produce more because of AI.

My advice: build it into the process

Build quality into how your team works with AI. Put serious time into the setup before AI starts writing, so it works from your facts, your sources and your voice. Then add multiple checks along the way, not only at the end.

This slows you down at first. After that, you're faster and safer with every piece.

→ Skip the setup and checks, and most errors that go out under your name are ones you could have prevented.

Yes, the setup slows you down, for a few weeks

You bought AI to go faster, and now someone tells you to slow down first. That's a hard sell when your AI investment has to pay back this year.

But that changes after the first weeks. You build the setup once and use it for every piece after. The more you publish, the less care each piece costs. 

If you only produce a handful of pieces, it flips. Then the setup costs more than it saves, and one careful check at the end is enough.

More than 70% of marketers already spend one to five hours a week fact-checking AI output. A team that checks everything by hand at the end pays the full price every time.

→ The setup is a cost once. Skipping it is a risk every time.

AI errors are predictable

AI doesn't fail at random. When AI content goes out with errors, it's often false facts or not exisiting source links. Behind those sit five predictable habits, AI:

• makes things up
• leaves things out
• uses outdated information
• answers a different question than the one you asked
• uses language your brand would never use

That's good news. What you can predict, you can check for. Not a vague "have a look before it goes out", but a concrete check for each of those habits.

→ Predictable errors are preventable errors.

Checking at the end is too late

Most teams check the content at the end, if they check at all. By then the error sits inside a finished piece and often goes unnoticed. If someone does catch it, fixing it costs time nobody planned.

Put the time in earlier. Configure AI before it writes: your facts, your sources, your voice. Build in checks and double-checks along the way. Then a third check at the end, just before publication.

Skip the checks and you get DPD, the parcel company whose chatbot swore at a customer. Or lawyers filing case law that AI made up. In both cases, nobody checked what AI did before it went out.

With my clients, no AI error has gone out. Not because AI doesn't make them. Because we double down on configuration and checking throughout the process.

Start here

Ask your team to walk you through one piece made with AI: from the setup, to the brief, to the checks, to publication. 

Then ask one question: does anyone asks AI to check and push back on what AI wrote, or do we only check the links? If it's only the links, the quality isn't built in. It's hoped for.

Fast and safe is a decision

Speed and safety can go together, in a process that’s built for both. Set it up once, use it often, and check your setup often, especially after every AI update. Without that process, an impossible deadline, speed expectations or too busy employees decides how accurate you are.

If you want to be fast and safe, but nobody knows how to set it up, I'll help you build the process. Email me directly.

9 AI dilemmas

This is one of nine AI dilemmas for the C-suite. I'm publishing them one by one over the coming weeks. Each piece names the decision, what it costs to leave it open, and what I'd do. 

The one you recognise first is usually the one you've left open the longest.

Sources

• NP Digital, AI Hallucinations and Accuracy Report (February 2026)
• HubSpot, State of Marketing 2026
• DPD chatbot incident (January 2024)
• AI hallucinations in court filings

Join 2,750+ professionals

AI changes how the work gets done. I test it in my own work and write about what actually works.

Free. Unsubscribe anytime.

→ Subscribe