LinkedIn wants less AI, gets cited by AI anyway, and checks your ad receipts

We’re going to give you a cake, but don’t even think about eating it.

The platform is pulling in two directions: Restricting AI-generated content that lacks a human perspective while building AI writing tools directly into its composer.

Global Editorial VP Laura Lorenzetti says: “When AI is overused, especially at scale, it dilutes the valuable insights that real human conversations can spark.” Hm…

This means AI-heavy LinkedIn content will reach fewer people, so your organic strategy needs a real human voice behind it, or it won’t travel.

AI loves LinkedIn, however: According to a Meltwater study of 9.5M citations, LinkedIn is the second-most-cited domain among major AI chatbots, particularly for B2B queries.

So, when a potential customer asks a chatbot about your industry, there’s a real chance LinkedIn content shows up in the answer. Yours could be one of them.

What gets cited:

  • Expert-led individual posts over company page updates.
  • Articles and plain text (83% of all citations).
  • Clear headings, numbered lists, and specific data points.

Push your internal experts to post. Structure everything with headers and bullets. Generic brand content won’t get cited.

Follow the paper trail: LinkedIn has added post-bid measurement for Audience Network campaigns through DoubleVerify.

Meaning, after your ads run, you get independent reporting on viewability, brand safety, fake clicks, and whether your ads actually showed up in the right places.

LinkedIn already filters placements before bids. Now it checks the receipts after.

If you’re spending outside the core feed, that means you can see exactly where your budget went and cut anything that isn’t pulling its weight.

Continue Your Journey

Buyers are making up their minds, so meet them mid-decision

19. June 2026
1 min read

Turns out the internet’s front page doubles as its fitting room. Lurkers before they leap: A fresh Reddit study with WPP surveyed 13,000+ users and found that 77% of Redditors hop across two or more communities before they buy. Why it matters: Most marketing playbooks chase awareness and buzz. But the actual yes-or-no moment happens…

AI Overviews may be favouring your competitors, and Customer Match goes auto-classified

19. June 2026
2 min read

Turns out being “seen” isn’t the same as being “chosen.” The plot twist nobody asked for: Google’s AI Overviews recommend your competitors 69% of the time, even when they cite your own self-promotional “best” listicle. Lily Ray studied 100 B2B “best [category] software” queries across three dates. Of the 80 that triggered an AI Overview,…

Facebook uses AI mode to answer user questions, Threads celebrates a milestone with new features

17. June 2026
2 min read

Turns out everyone wants to be the place you ask questions. Ask, and ye shall receive…a Meta AI answer: Facebook launched AI Mode in search and now answers questions with AI-generated responses. These outputs are pulled from public posts, Groups, and Reels instead of the usual link list. “Real answers from real people,” as Meta…