What 80,000 People Want From AI — And What It Means for Your Business
Anthropic's landmark study of 80,508 people across 159 countries reveals a surprising truth: people don't want AI to work faster. They want to live better. Here's what that means for small businesses.
Anthropic — the AI company behind Claude — just published the largest multilingual qualitative AI study ever conducted. They interviewed 80,508 people across 159 countries in 70 languages and asked a deceptively simple question:
What do you actually want from AI?
The answer wasn’t “make me faster.” It wasn’t “generate more content.” It wasn’t even “automate my job.”
People don’t primarily want AI to work faster — they want to live better.
— The study’s central finding
The dominant theme across 80,000 responses: take the busywork off my plate so I can focus on what matters. That’s not an AI story. That’s a human story. And for anyone running a small business, nonprofit, or professional services firm, it’s deeply personal.
The Study at a Glance
Before we dig in, here’s what makes this research exceptional:
| Detail | Value |
|---|---|
| Participants | 80,508 Claude users |
| Countries | 159 |
| Languages | 70 |
| Method | Conversational AI interviews (not surveys) |
| Duration | One week in December 2025 |
| AI delivery rate | 81% said AI already took steps toward their vision |
| Avg. concerns per person | 2.3 — people hold hope AND worry simultaneously |
| Global sentiment | 67% net positive |
This isn’t a poll with five answer choices. These were open-ended conversations where people described their actual experiences, frustrations, and hopes in their own words. The depth of the responses is what makes the findings so valuable.
What People Actually Want
The study categorized responses into nine distinct visions for AI. The top five paint a clear picture:
| What People Want | % of Respondents | In Their Words |
|---|---|---|
| Professional Excellence | 18.8% | Handle routine tasks so I can focus on strategic, meaningful work |
| Personal Transformation | 13.7% | Help me grow, understand myself, achieve wellbeing |
| Life Management | 13.5% | Reduce the mental burden of managing everything |
| Time Freedom | 11.1% | Give me my time back — for family, hobbies, rest |
| Financial Independence | 9.7% | Help me achieve economic security |
The remaining categories — Societal Transformation (9.4%), Entrepreneurship (8.7%), Learning & Growth (8.4%), and Creative Expression (5.6%) — round out the picture.
Here’s what jumps out: roughly one-third of all respondents want relief from time, money, and mental bandwidth constraints. Not more features. Not more capabilities. Relief.
“I didn’t want AI to make me a better worker. I wanted it to give me back the hours I was losing to tasks that shouldn’t need a human.”
— Paraphrased from study respondents
What’s Missing From This List
Notice what nobody said:
- “I want a chatbot”
- “I want to generate images”
- “I want a smarter search engine”
The asks are deeply operational and deeply human. People want to stop doing busywork, stop feeling overwhelmed, and start spending time on what actually matters to them — whether that’s growing a business, serving a community, or being present with family.
Where AI Has Already Delivered
The study didn’t just ask what people want. It asked what AI has actually done for them. The results are encouraging — and revealing:
| Where AI Delivered | % | What It Looks Like |
|---|---|---|
| Productivity | 32% | Dramatically accelerated work, automated repetitive tasks |
| Cognitive Partnership | 17.2% | Served as thinking partner and creative collaborator |
| Learning | 9.9% | Helped acquire new skills and understanding |
| Technical Accessibility | 8.7% | Enabled building things that were previously impossible |
| Research Synthesis | 7.2% | Processed large volumes of information efficiently |
| Emotional Support | 6.1% | Provided guidance and judgment-free space |
But there’s a notable gap: 18.9% said AI fell short of their expectations. The technology has delivered real value, but it hasn’t yet closed the distance between what people want and what they’re getting.
That gap is the opportunity.
Why This Matters More for Small Businesses
Here’s where the data gets really interesting for anyone running a small or mid-size organization.
Independent Workers Are 3x More Empowered
The study found a striking occupational pattern: independent workers — entrepreneurs, freelancers, small business owners — report 3x higher AI empowerment than employees at larger organizations (47% vs 14%).
Why? Because when you’re running a small business, every hour of busywork is an hour you’re not:
- Growing the business
- Serving clients
- Building donor relationships
- Being present with your family
AI doesn’t just save time for independent workers — it fundamentally changes what’s possible. A solo operator with the right AI tools can handle operational complexity that previously required a team.
The Equalizer Effect
The study also found that AI optimism peaks in emerging markets — South America, Africa, South and Central Asia — where people see AI not as a threat but as an opportunity ladder.
One respondent, an entrepreneur from Cameroon, put it perfectly:
“I’m in a tech-disadvantaged country… AI did it in 30 seconds. It’s an equalizer.”
This resonates deeply with us. The organizations most underserved by technology — small businesses running on spreadsheets, nonprofits juggling disconnected tools, professional services firms drowning in manual data entry — are exactly the ones that stand to gain the most from purpose-built AI.
It’s also why we architected Muin for global reach from day one — multi-language support (English, Spanish, Arabic with RTL), multi-currency (33 currencies), regional data residency routing so your data stays in your geography, and WhatsApp as a first-class communication channel alongside email and SMS. When an entrepreneur in Cameroon says AI is an equalizer, we take that seriously. The infrastructure to serve them shouldn’t be an afterthought — it should be in the foundation.
Entrepreneurship as Capital Bypass
The study found that 8.7% of respondents see AI as a force multiplier for building businesses — and this vision peaks in Africa, South/Central Asia, the Middle East, and Latin America. People in these regions frame AI as a way to bypass the capital and infrastructure barriers that have historically kept small organizations from competing with larger ones.
That’s not a technology story. That’s an economic mobility story. And it demands software that’s ready to meet them where they are — in their language, their currency, and their region.
The Trust Gap: People Want AI But Don’t Trust It Yet
The study identified 13 categories of concern. People aren’t naive about AI’s risks — on average, each respondent mentioned 2.3 distinct concerns. The top five:
| Concern | % | The Worry |
|---|---|---|
| Unreliability | 26.7% | Hallucinations, inaccuracy, burden of verifying AI output |
| Jobs & Economy | 22.3% | Displacement, unemployment, growing inequality |
| Autonomy & Agency | 21.9% | Loss of human control, forced adoption |
| Cognitive Atrophy | 16.3% | Losing skills, intellectual passivity, critical thinking decline |
| Governance | 14.7% | Lack of legal frameworks, insufficient oversight |
Other significant concerns include misinformation (13.6%), surveillance and privacy (13.1%), malicious use (13%), and meaning/creativity loss (11.7%).
The #1 Concern Is Solvable
Unreliability — hallucinations and inaccuracy — tops the list at 26.7%. It’s also the only “tension” in the study where the negative experience outweighs the positive.
Here’s the thing: this concern is largely driven by general-purpose AI tools that are designed to do everything and therefore improvise on everything. When you ask a general chatbot to process an invoice, classify a compliance document, or route an approval workflow, you’re asking it to wing it.
Purpose-built AI — AI designed for specific business tasks within defined guardrails — is fundamentally more reliable than general-purpose chat. Structured extraction pipelines with validation, confidence scoring, and human-in-the-loop review for edge cases are different in kind from “ask the AI and hope for the best.”
The study validates what we’ve seen firsthand: the path to AI trust isn’t better chatbots. It’s AI embedded into business operations where it can be verified, validated, and held accountable.
Privacy Is a Top Concern
13.1% of respondents worry about surveillance and data exploitation. This concern is highest in North America (17%) and Oceania — the very markets where most AI tools are built.
For small businesses handling sensitive data — financial records, employee information, donor details, client contracts — this concern is not abstract. It’s operational.
The architecture of your AI matters. Where your data goes, who can access it, and whether it’s used to train models that benefit someone else — these aren’t philosophical questions. They’re business decisions.
At Muin, we chose AWS Bedrock specifically because of its contractual guarantees: your data is never used to train AI models, and it never leaves our secure infrastructure. That’s not a policy that could change — it’s architecture.
Five Tensions: What the Study Reveals About Living with AI
The most nuanced finding in the study is that benefits and risks aren’t experienced by different people — they’re experienced by the same people simultaneously. The researchers identified five key tensions:
1. Learning vs. Cognitive Atrophy
- 33% mentioned AI helped them learn; 17% worried it would erode their skills
- 91% of those who cited learning benefits had experienced them firsthand
- Only 46% of those worried about atrophy had actually witnessed it
What this means for business: AI should automate operational tasks (data entry, routing, classification) — not judgment tasks (strategy, relationships, decisions). The goal is to free up human capacity, not replace human thinking.
2. Better Decision-Making vs. Unreliability
- 22% valued AI for decision support; 37% worried about reliability
- This is the only tension where the negative outweighs the positive
- Lawyers report both the highest benefits AND the highest reliability concerns (nearly 50%)
What this means for business: AI-generated insights need to show their source data. Confidence scores, citation trails, and human approval workflows are essential — not optional.
3. Time-Saving vs. Illusory Productivity
- 50% cited time savings; 18% worried about losing time to AI or accelerating pace demands
- Self-employed workers are most likely to experience both sides simultaneously
- Only 1% have actually experienced illusory productivity — 17% just expect it
What this means for business: Real AI productivity doesn’t just speed up tasks — it eliminates entire categories of work. You don’t process invoices faster; the invoices process themselves. That’s the difference between “faster treadmill” and “getting off the treadmill.”
4. Economic Empowerment vs. Displacement
- 28% cited economic benefits; 18% feared displacement
- Independent workers report economic benefit at 3x the rate of institutional employees
- Freelance creatives experience the tension most acutely: 23% benefit, 17% displacement
What this means for business: AI platforms should empower small organizations to compete — giving a 5-person team the operational capacity of a 50-person organization. That’s empowerment, not displacement.
5. Emotional Support vs. Dependency
- 16% found meaningful emotional support from AI; 12% feared dependency
- People who value emotional support are 3x more likely to also fear dependency
- A Ukrainian soldier noted that AI companions helped during wartime when human support was unavailable
What this means for business: AI in business should be a tool, not a companion. It manages your operations so you have more time for human relationships — not less.
The Global Picture: AI Optimism by Region
One of the study’s most fascinating findings is geographic. AI sentiment varies dramatically by region:
| Region | Net Sentiment | Key Driver |
|---|---|---|
| South America | Most positive | AI as economic opportunity |
| Sub-Saharan Africa | Most positive | AI as equalizer; 18% expressed no concerns at all |
| South & Central Asia | Highly positive | Learning vision 13-14% (vs 8% globally) |
| East Asia | Positive with nuance | Personal transformation peaks at 19% |
| North America | Mixed | Highest governance (18%) and surveillance (17%) concerns |
| Europe | Most cautious | Highest overall concern rates |
Why the divergence? In emerging economies, AI is framed as an opportunity ladder — a way to bypass the capital, infrastructure, and institutional barriers that have historically kept small organizations from accessing enterprise-grade capabilities. In developed economies, AI is more often framed as a threat to established systems and employment.
This maps directly to the SMB opportunity. For organizations that have been underserved by enterprise software — whether a growing business in Lagos, a nonprofit in Louisville, or a professional services firm in London — an AI-powered platform isn’t a disruption. It’s an invitation to finally compete on a level playing field.
What This Means for Business Software
The Anthropic study confirms something that many of us in the small business world have felt intuitively but couldn’t prove with data:
The AI products that will win are the ones that deliver operational relief — not novelty.
People don’t want another tool to learn. They want fewer tools that do more. They want their weekends back. They want to stop worrying about the grant report that’s due, the invoice that was processed wrong, the compliance certification that expired without anyone noticing.
That’s what we’re building at Falaah AI with Muin — and the study data maps directly to the capabilities we’ve built.
From Study Findings to Real Capabilities
Here’s how each finding connects to what Muin actually does:
“Handle routine tasks so I can focus on strategic work” (18.8%)
This is the job of Muin Agents — AI-powered workers that handle repetitive operations 24/7 without human intervention. The Invoice Agent reads, extracts, validates, and books invoices automatically. The Compliance Agent monitors regulatory changes and flags policy gaps before they become audit findings. The Certification Agent tracks expiration dates across your entire team and sends multi-tier alerts before deadlines hit.
These aren’t chatbots. They’re purpose-built AI workers that do the work — not just answer questions about it.
“Reduce the mental burden of managing everything” (13.5%)
The mental burden comes from juggling disconnected tools. QuickBooks for accounting. Gusto for payroll. Google Drive for documents. Spreadsheets for everything else. None of them talk to each other — and you’re the human integration layer holding it all together.
Muin connects and unifies that patchwork into a single platform: Document Intelligence for reading, extracting, and creating documents. Smart Forms for data capture with conditional logic and workflow triggers. Integration Hub to sync with the tools you already use (HubSpot, QuickBooks, Google Drive, Gusto, and more). One login, one source of truth, one place to look.
“Give me my time back” (11.1%)
This is the difference between faster and gone. Workflow Automation doesn’t just speed up your approval chains — it runs them automatically. A visual builder for multi-step processes: when an invoice arrives, it’s extracted, matched to a PO, routed for approval, and synced to your books. When a certification expires, the renewal workflow triggers. When a new hire signs their offer letter, onboarding kicks off across HR, payroll, documents, and training — without anyone staying late to set it up.
The study found that only 1% of people have experienced illusory productivity from AI — the worry that AI just makes you faster on a treadmill that keeps speeding up. That’s because the best AI doesn’t make you run faster. It takes you off the treadmill entirely.
“AI as a force multiplier for building my business” (8.7%)
The Muin Assistant lets you chat with your business data in natural language. “Which vendors have expiring insurance certificates AND outstanding invoices over $5,000?” — a question that’s impossible to answer when your data lives across six tools, but instant when it’s unified. Smart Payments lets you accept cards, NFC, ACH, and Apple Pay (via Stripe) — and send payments via Stripe Connect — with a transparent 0.5% platform fee. No stitching together three payment tools. No reconciliation nightmares.
For a solo operator or a small team, these capabilities mean competing with organizations ten times your size — without ten times the staff.
“AI as an equalizer” (peaks in emerging markets)
The Communications Hub unifies WhatsApp, email, and SMS into one inbox — with AI designed to draft and handle routine responses automatically. A congregant sends a WhatsApp message asking about donation receipts? AI responds with the right information, in the right language, instantly. A vendor emails asking about payment status? AI drafts the response from real data — you just review and send.
For organizations where staff time is the scarcest resource — a small nonprofit, a growing professional services firm, a house of worship with one administrator — AI-powered communications aren’t a luxury. They’re the difference between responsive and overwhelmed.
Addressing the Concerns
The study’s top concerns map to deliberate architectural decisions in Muin:
| Concern (Study) | How Muin Addresses It |
|---|---|
| Unreliability (26.7%) | Purpose-built AI for specific tasks — structured extraction, confidence scoring, human approval workflows. Not a general chatbot improvising on your business data. |
| Privacy (13.1%) | Built on AWS Bedrock — your data never trains AI models, never leaves our infrastructure. Full audit logs. 256-bit encryption. Privacy-first by architecture, not by policy. |
| Autonomy (21.9%) | Human-in-the-loop on every decision that matters. AI handles the operational work; you keep control of strategy, relationships, and judgment calls. |
| Cognitive Atrophy (16.3%) | Muin automates data entry, routing, and classification — not thinking. Your expertise stays sharp because you’re using it on the work that actually requires it. |
This isn’t a generic AI tool hoping to be useful. It’s a connected platform with business modules for Finance, HR, Compliance, and Non-Profits — with more on the way — where every capability is designed to deliver the exact operational relief that 80,000 people just told Anthropic they’re looking for.
The Bottom Line
Anthropic’s study asked 80,000 people what they want from AI. The answer was remarkably human:
Help me focus on the work that matters. Give me my time back. Take the busywork off my plate.
Not “make me a better prompter.” Not “generate more stuff.” Just: let me do less of what drains me and more of what fulfills me.
For small business owners, nonprofit leaders, and professional services partners, that’s not an abstract wish. It’s the difference between leaving at 5:30 and leaving at 8:00. Between growing the practice and just surviving. Between serving the community and drowning in paperwork.
The technology to deliver on that wish exists today. The question is whether it gets built for the organizations that need it most — or only for the ones that can afford six-figure implementation budgets.
We know which side we’re building for.
Sources
- Anthropic — “What 81,000 People Want from AI” — Full study with methodology, findings, and interactive visualizations
- AWS Bedrock Data Protection — AWS Bedrock’s data privacy and protection guarantees
- Salesforce 2025 SMB Trends Report — AI adoption trends among small and mid-size businesses
Related Reading
- The Small Organization Technology Gap — Why SMBs and nonprofits are being left behind by AI
- How a 5-Person Business Can Operate Like a 50-Person Company — Before/after story with four SMB archetypes
- Why Privacy-First AI Matters for Your Business Data — Why architecture matters more than policy
- How a 15-Person Non-Profit Can Run Like a 150-Person Org — The nonprofit-specific version