How 81,000 People Feel About AI - And How Your AI Strategy Can Benefit From It
Anthropic just published what may be the largest qualitative study on AI ever conducted - 81,000 people sharing their hopes, fears, and lived experiences with AI.
Anthropic’s groundbreaking study is worth reading in full (link below). What struck us most wasn’t the optimism or the anxiety in isolation - it was how tightly the two are bound together in the same person. People want AI to free up their thinking. They’re also afraid it’ll replace it. We’ve felt those exact tensions while building our AI legal company, Librari. Here’s how we’re thinking about them.
The professionals who stay ahead already know what they agreed to.
The #1 thing people want from AI is professional excellence: handling the routine so they can focus on higher-value, strategic work. For legal, finance, and procurement teams, that routine is contract review. Searching drives. Checking renewal dates. Pulling terms from PDFs.
Librari does that work for you - so you can make the decisions.
“If AI truly handled the mental load… it would give me back something priceless: undivided attention.” - Manager/Executive, Denmark
That’s the shift. Not faster busywork. Actual thinking time on the decisions that matter.
What Librari gives back:
Better decisions. 22% of people in the survey cited AI-assisted decision-making as a tangible benefit they’d already experienced. For contract teams, that means surfacing the right clause before the negotiation, not after the dispute.
Time back. Half of all respondents cited time savings as a real benefit they’d lived. For contract-heavy teams, that time compounds fast.
Less mental load. 14% of respondents specifically wanted AI as cognitive scaffolding - managing complexity so they could focus on what actually requires their judgment. That’s exactly what Librari does with your contract portfolio.
You’ve been burned by AI before. We designed for that.
Unreliability is the single most common concern in AI adoption: 27% of people worry about hallucinations, inaccurate outputs, and the burden of having to verify everything the AI tells them. Nearly half of all lawyers specifically reported coming up against AI unreliability firsthand.
We know. We built Librari for exactly those people.
“An assistant that sounds sure but is often wrong forces you to treat everything as suspect. Instead of freeing attention, it creates a permanent ‘fact-check tax.’” - United States
Here’s why contract intelligence is different:
Contracts are deterministic. A renewal date is a renewal date. A payment term is a payment term. There’s no “interpretation.” Librari extracts what’s there. When the data is in the document, it’s right.
We built Librari Verify. Our new continuous extraction evaluation system runs ongoing accuracy checks against your contract data. We know where we’re strong. We know where we’re getting stronger. And we tell you.
We never guess. Where Librari doesn’t have high confidence in an extraction, it says so. We’d rather surface uncertainty than manufacture a wrong answer.
We designed Librari to be the opposite of that fact-check tax: precise where precision is possible, and honest where it isn’t.
Librari doesn’t replace your legal team. It makes them better and faster.
22% of people surveyed cited job displacement as a primary concern - the strongest single predictor of negative AI sentiment overall.
If you’re a lawyer or paralegal reading this: we hear you. And we think the concern is aimed at the wrong target.
The threat to legal work isn’t contract intelligence. It’s the team that has contract intelligence while yours doesn’t. They catch the auto-renewal before it locks. They know the liability cap before the negotiation starts. They move faster, with more confidence, on every deal.
Workers consistently said they wanted AI to automate routine tasks while preserving the parts of their jobs that define their expertise. That’s exactly the split Librari makes. We handle the finding, the organizing, the surfacing. You handle the thinking, the judgment, the strategy.
The goal isn’t fewer lawyers. It’s better ones.
Your contracts don’t leave your environment. Ever.
This one didn’t make the Anthropic survey - but it’s the first thing our users ask us. So we’ll say it plainly.
Most AI tools move your data to analyze it. Librari doesn’t. Your contracts stay in your systems. We connect to them, read them, and surface intelligence. The files never transfer to us.
What that means practically:
Your Google Drive stays your Google Drive.
Your DocuSign vault stays your DocuSign vault.
Librari sees what you authorize. Nothing more.
Role-based access, respected. If someone in your company can’t see a contract in your existing system, they can’t see it in Librari either. We honor the permissions you’ve already set. Finance sees finance. Legal sees legal. No cross-contamination.
CASA certified. SOC 2 Type II aligned and certification in progress.
The hesitancy is the problem.
The contracts you’re not tracking right now are still active. Those renewal clauses are still running. Those liability terms are still in effect. Not connecting Librari doesn’t make your contract exposure go away. It just keeps it invisible.
81% of people who engaged with AI said it had already taken a meaningful step toward their stated goal. The gap between “I’ve heard about this” and “this is actually working for me” is smaller than you think.
Connect in 60 seconds. No migration. No long-term commitment. No data movement.
Start free - connect your first integration
Quotes and statistics sourced from Anthropic’s 81K Survey: “What 81,000 People Want from AI,” published March 18, 2026. Full study at anthropic.com/features/81k-interviews.