For decades, Brussels associations have treated the white paper as the ultimate credibility play: serious, evidence-led, sometimes months in the making.

The uncomfortable truth? Many of them barely got read. Skimmed for a few quotes, then quietly filed away.

Meanwhile, the rise of short-form platforms and “everything in 30 seconds” culture led some to predict that policy communication would become a stream of ultra-short content, mainly video. Possibly with dance moves. Brussels, mercifully, didn’t fully follow that script.

So white papers stayed relevant, but practitioners adapted. They distilled the substance into briefings, talking points, social snippets, and the usual Thursday ritual in Place Lux, where a quick coffee becomes a strategy session (see my thoughts on going Beyond Coffee).

Now AI is changing discovery again. And this is where it’s worth being precise.

It’s tempting to say “AI tools search the web,” but the reality is more nuanced. Some systems answer purely from what the model learned during training, a static snapshot of the internet. Others use retrieval-augmented generation: they pull live results from the web or an index, then synthesise an answer from what they find. Google’s AI Overviews and Perplexity work this way. The distinction matters, because it changes what content actually influences what people see, and when.

So what does this mean for white papers?

Not that PDFs magically become the winning format. Not that “more pages” equals more visibility. But there’s a reasonable case that in an answer-first world, high-quality, well-structured evidence has a better chance of being surfaced, cited, and reused. Especially when it’s accessible and easy to corroborate.

Here’s the practical shift: if your strongest thinking lives only in a 60-page PDF that’s hard to parse, hard to find, and never linked, you’re leaving influence on the table. Publish that same thinking in a crawlable, structured format, with clear claims and modular sections, and it becomes far easier for both humans and AI systems to pick up and reuse.

Discoverability isn’t just about format, of course. It also depends on whether your content lives on a domain that AI systems already trust and crawl frequently. A beautifully structured page on an obscure subdomain may still struggle to surface.

So maybe the white paper gets a second life. Not as a monolith, but as a content spine. Something that can be read in depth by the people who need it, while also feeding smaller, linkable pieces that travel.

My current working view:

• Do the deep work. Quality still beats noise.
• Publish in formats that travel. HTML hub page first, PDF as download.
• Make verification easy. Clear sources, data, references.
• Build in modularity. FAQs, standalone sections, reusable explainers.

The white paper might not return exactly as it was. But the underlying need for credible, structured, evidence-based argumentation looks like it’s becoming more valuable, not less.

Curious if you’re seeing the same thing. Are your long-form policy assets becoming more useful again, or just being repackaged into smaller formats?

This article was originally posted by Philip Weiss on LinkedIn