Here is something that took me an embarrassingly long time to figure out about myself.

I can listen to a six-hour audiobook and recall it in detail a week later. I can sit in a complex conversation, track every thread, and synthesise what matters in real time. I can consume hours of audio content without losing focus once.

But hand me a long memo — something that genuinely matters, something I actually want to understand — and something in my brain quietly refuses to cooperate. Not out of disinterest. Not out of laziness. Out of something I couldn't name for years.

The information was the same. The format was different. And for reasons I didn't fully understand, the format was everything.

It wasn't until much later — after I'd built years of workarounds without knowing why I needed them — that I understood what was actually happening. I was eventually diagnosed with mild ADHD. Not the kind that announces itself loudly. The quieter, inattentive kind that looks, from the outside, like distraction or disengagement, but feels, from the inside, like a mind searching desperately for the context it needs before it can hold the detail.

I am far from alone. A landmark 2024 meta-analysis estimates that ADHD affects an estimated 366 million adults worldwide — and less than 20% have received a formal diagnosis or are currently being treated. Most of them, like me for many years, have simply been told — implicitly or explicitly — that the problem is them.

It isn't. It never was.

The memo problem — and how AI solved it.

Here is what information overload actually feels like from the inside.

You open a long document — a memo, a research paper, an article — and you know it matters. You want to engage with it. But before you can find the entry point, before you have any sense of the shape of the thing, you are already being asked to process the first sentence of a structure you cannot yet see. The context hasn't arrived yet. The relevance hasn't been established. And without that — without knowing why this matters and where it's going — the mind has nothing to hold the attention in place.

So you lose the thread. Sometimes in the first paragraph. Sometimes in the second. And then you either read the whole thing again with diminishing returns, or you walk away having absorbed very little of something that genuinely mattered.

I spent years developing workarounds. Reading the last paragraph first. Skimming for headings. Asking colleagues to summarise before I committed to the full read. None of these were ideal. All of them were ways of hacking a format that wasn't designed for how I think.

Then AI arrived. And something changed.

Now, before I engage with a long document, I ask AI to give me the shape of it first. The core argument. The three things I need to understand before the detail makes sense. The one question this document is actually trying to answer.

And then something interesting happens: I want to read it.

Because now I have context. I have a scaffold to hang the detail on. The interest has been cultivated before I've committed to the linear experience. And when I do go back to read specific sections in depth — the parts that need elaboration, the nuances that matter — I'm reading with comprehension rather than against resistance.

A 2024 study in the Journal of Attention Disorders found that multi-format information presentation improved retention in ADHD participants by 41% compared to text-only formats. What AI enables is precisely this — the ability to receive information in the format that fits how you think, rather than the format that was convenient for whoever wrote it.

The TL;DR is not a shortcut. For millions of people, it is the door.

This is not a productivity story.

I want to be precise about what this is and isn't.

This is not about doing more faster. This is not about optimising output or squeezing efficiency from a reluctant mind.

This is about access.

Reading is one of the most cognitively demanding tasks we ask people to do. It requires the simultaneous use of multiple skills — auditory processing, working memory, processing speed, and long-term memory — all operating at once. For people whose working memory and processing speed don't operate the way the system assumed they would, reading long-form text is not just difficult. It is a sustained act of cognitive labour that exhausts the very resources needed to comprehend what's being read.

Audiobooks work for me because they remove the decoding layer. The format does the sequential work. My mind can focus on meaning rather than on the mechanical act of processing symbols on a page.

AI summaries work for the same reason — but they go further. They don't just change the delivery format. They give the mind what it needs to engage: context first, detail second, question-and-answer as a navigation tool rather than a linear march through someone else's structure.

Research analysing AI tools and executive function has identified four primary mechanisms by which generative AI addresses these challenges: cognitive scaffolding that compensates for executive function deficits, task decomposition that breaks complex information into manageable components, real-time contextual support that reduces attention-switching costs, and personalised systems that adapt to individual patterns.

In plain language: for the first time, information is becoming format-agnostic. The content exists. The format adapts. And that adaptation — small as it sounds — is genuinely life-changing for minds that have spent decades being failed by format, not by capacity.

What this tells us about intelligence and systems.

Here is the thought that stays with me.

The people who struggled to read long documents linearly were never less intelligent. They were not less curious, less capable, or less deserving of the information inside those documents. They were simply operating in a system that had decided one format was the default — and concluded that anyone who couldn't navigate that format was the problem.

For adults whose attention works differently, consuming information through audio or structured summaries can dramatically reduce the cognitive load associated with decoding text — freeing mental energy to focus on content and meaning rather than the mechanics of processing. This has always been true. The technology to do it at scale — across every document, every memo, every research paper, every article — simply didn't exist until now.

AI didn't create new intelligence. It removed a barrier that was preventing existing intelligence from being expressed.

That is not a small thing. For hundreds of millions of people who have spent their lives being failed by format, being told to try harder, to focus more, to be more like everyone else — it is everything.

The problem was never the mind.

It was what we asked it to do, and how we asked it to do it.