The Marketing edition
The AI news of that week, in plain language.
Platforms are filtering out AI junk, PwC blundered with made-up sources, and from now on the AI Act asks for honest labels. This week is about one thing: credibility.
PwC blunder: AI reports full of made-up sources cost the firm its reputation
Four advisory reports from PwC turned out to be written by AI, with studies that do not exist and made-up quotes. This is exactly the trap you can fall into too.
Four reports from PwC Middle East, published between 2024 and 2026, turn out to be largely written by AI. They were full of made-up footnotes, MIT studies that do not exist, and fabricated quotes. A detection company found it, and the Financial Times confirmed it. The reports were meant as 'thought leadership' to win new clients. Exactly the kind of content marketers make.
Background for subscribersDo you let AI help write a blog, whitepaper, or newsletter with facts in it? Treat every text as a first draft from an intern. Look up every figure ("67% of customers…") and every source in the original. If you cannot find it, it was made up. Your brand name goes on it, not the tool's. One unchecked number that a client checks costs you exactly what PwC lost here: trust.
LinkedIn gives readers a button to call your post 'AI junk'
Since 31 July, LinkedIn users can report posts as 'Seems like AI slop'. Such posts get less reach. The reason: 41% of longer LinkedIn posts turn out to be made entirely by AI. LinkedIn is even scrapping its own 'Enhance your post' writing aid and replacing it with a spellchecker that keeps your own voice.
Background for subscribersThis is a taster.
Subscribers read the whole Marketing edition: every story, each with its background in plain language.
Subscribe →