Not because you're lazy. Because you've done the math: two hours to read it, ninety minutes to absorb it, and somewhere inside — probably on page 41 — is the one paragraph that actually changes what you do next. The rest is context you already have, caveats you don't need, and tables that exist to justify the report's existence.
You're not behind. You're making a rational call. The reading cost doesn't justify the return. So the report sits. The insight stays buried. And you make your next decision without it.
AI changes that calculation.
The Problem Isn't Reading Speed. It's the Cost Model.
Knowledge workers in their 40s and 50s are processing more written material than ever — and they're doing it at the exact moment in their careers when time is most expensive. Research papers, regulatory filings, consultant decks, industry analyses, vendor comparisons: the volume hasn't slowed, and the expectation that you've absorbed all of it hasn't either.
The traditional solution was to get faster at reading. Skim the executive summary, cherry-pick the charts, flag it for a future read you'll never schedule. The result is a permanent stack of half-digested information and a nagging sense that somewhere in there is something important.
AI changes this equation. Instead of reading the document, you interrogate it. You ask the questions you actually need answered, and the AI pulls the relevant passages. You stay in command of the analysis — you just skip the fifty pages that don't touch your decision.
Document synthesis is one of the most immediately practical AI skills you can use today.
The Skill: AI Document Interrogation
Two tools do this particularly well in 2026.
Claude is built for single-document interrogation. Upload a PDF — a 60-page report, a lengthy contract, a dense research paper — and then ask targeted questions: What are the three main recommendations? What risk factors are flagged in section four? What does this say about implementation timelines? Claude reads it, surfaces the relevant sections with references, and gives you a pointed answer. You can follow up, push back, or drill deeper. The document becomes a conversation partner.
NotebookLM handles multi-document synthesis. Where Claude works best on a single source with focused questions, NotebookLM lets you feed in multiple sources simultaneously — upload five reports, three articles, and two internal documents — and then query across the whole collection. It answers in the context of your source material and cites exactly where each answer came from. For research-heavy work or any situation where you're synthesizing across a body of evidence, it's the right tool.
Both require no technical setup beyond creating an account.
How to Do It: The 15-Minute Document Interrogation
Take any report you've been avoiding. Here's the process.
With Claude (claude.ai):
Start a new conversation and click the paperclip icon to upload your PDF.
Don't start with "summarize this" — that returns a generic overview shaped by the document's own framing, not your priorities. Ask your actual question: What does this say about [specific topic]? or What are the three things I need to act on?
Follow up. Ask Claude to show you where it found the answer. Ask what the document says about counterarguments. Ask what the biggest flagged risk is. Keep the questions specific to your decision.
End with: Is there anything in this document I should know that I haven't asked about? — that's your catch-all for blind spots.
Total time to extract actionable insight from a 60-page document: 10–15 minutes.
With NotebookLM (notebooklm.google.com):
Create a new notebook and add your sources — PDFs, URLs, or Google Docs.
Ask synthesis questions across your sources: What do these reports agree on? Where do they diverge? What finding is most consistent across all of them?
Use the Discover panel to surface automatically identified themes and key quotes with source citations.
Total time to synthesize five sources into a coherent picture: 15–20 minutes.
The Output Gain
Here's what actually changes.
Before this skill: a 60-page report represents a 2–3 hour commitment to read, absorb, and extract the decision-relevant material. Most professionals skip it — they skim the executive summary, defer the full read, and move on with incomplete information.
After: the same report is a 15-minute interrogation. You get the three decisions buried inside it, with source references you can verify. You don't lose the depth — you skip the parts that don't touch your work.
Compounded across a week, a month, a year: this is the difference between being current on your field and being buried under it. Professionals who build this habit move through their reading load faster and arrive at decisions with more complete information.
That's not incremental. It's a fundamentally different relationship with the information you're supposed to be on top of.
One Thing to Try This Week
Find one document in your backlog — something you've been meaning to read but haven't. A report. A research paper. A contract or policy document.
Upload it to Claude or create a NotebookLM notebook. Spend 15 minutes asking three specific questions about what you actually need to know from it.
See what you find.
You've been carrying that document around in the back of your mind. Fifteen minutes from now, you can close the loop on it.
GenXRewired publishes practical AI skills for professionals with experience. If someone in your network is still reading every document front to back, forward this to them.