Americans Are Using AI More and Still Worried About It
I use AI often enough to understand the attraction: a good model can turn a blank page into a working draft, explain unfamiliar code, summarize research and remove hours of repetitive work.
But there is a strange feeling surrounding the technology right now.
The more useful AI becomes, the harder it is to say that Americans are becoming comfortable with it.
New Gallup research commissioned by Microsoft finds that 68% of Americans who use AI daily still say the technology makes them feel worried. The number falls among less frequent users, but worry remains the dominant emotion for a large share of the public.
That gap between use and trust is becoming one of the most important stories in AI.
The Headline Number Is 68%, But the Story Is More Complicated
The most attention-grabbing figure from the new Gallup research is 68%.
That is the share of U.S. daily AI users who reported feeling worried about AI. Among Americans who use AI less frequently, the figure rises to 80%, while 74% of those who have never used AI report feeling worried.
So experience appears to matter.
But it does not erase the concern.
Gallup explicitly cautions that the survey cannot determine whether people become more trusting because they use AI more often, or whether people who already trust AI are more likely to use it frequently.
America Is An AI Adoption-and-Anxiety Outlier
Gallup surveyed AI-aware adults across 37 countries and found a striking regional pattern.
Wealthy Western countries reported some of the highest levels of AI worry, while also having relatively high rates of AI use. The United States was the highest-worry country in the current group, with 74% of AI-aware adults saying AI makes them feel worried.
Only 36% of U.S. respondents expected AI to mostly help their country.
That creates an important distinction between personal usefulness and national optimism. Someone can use ChatGPT to finish a report faster and still believe AI could create serious problems for society.
Pew Finds a Similar Trust Problem
Gallup is not the only major survey showing American caution.
A June 2026 Pew Research Center survey found that 52% of U.S. adults are more concerned than excited about the increasing use of AI in daily life. Just 9% said they were more excited than concerned, while 37% said they felt equally concerned and excited.
The distinction matters because Pew's question measures a different thing from Gallup's emotion question. The two surveys should not be treated as interchangeable, but they point in the same general direction: AI is increasingly familiar without becoming universally reassuring.
Americans Trust AI More When They Actually Use It
There is another number hiding behind the worry headline.
In Gallup's U.S. data, 45% of daily AI users said they trust AI a lot or completely to provide accurate information.
That compares with 26% among weekly users, 15% among people who use AI less than weekly and just 8% among people who have never used it.
The relationship is strong, but it does not prove causation.
Trust and Accuracy Are Different Questions
A person may trust AI enough to brainstorm, summarize or rewrite a paragraph without trusting it to provide authoritative facts.
That is increasingly how responsible AI use works: the user trusts the tool for some jobs and verifies it aggressively for others.
Jobs Are Driving a Large Part of the Anxiety
Americans are becoming more concerned about AI's effect on employment as the technology moves from demonstrations into ordinary workplaces.
Pew found in June 2026 that 71% of U.S. adults think AI will lead to fewer jobs in the United States over the next 20 years, up from 64% in 2024.
Only 5% believe AI will lead to more jobs, while 10% think it will not make much difference.
For younger workers, the concern is particularly relevant. Among adults ages 18 to 29, 73% said AI would lead to fewer jobs, compared with 61% in 2024.
“Success in creating AI would be the biggest event in human history.”
— Stephen Hawking, writing in 2014 about the potential benefits and risks of artificial intelligenceHawking's warning was made long before today's generative-AI boom, but the underlying question remains familiar: how do we capture the benefits of increasingly capable machines while managing the risks that come with them?
Privacy May Be Even More Immediate Than Job Loss
One of the strongest findings from Pew's 2026 U.S. research concerns personal information.
71% of Americans say increased use of AI will make their personal information less secure. Only 3% expect AI to make their information more secure.
That is not an abstract concern.
Generative AI increasingly touches emails, documents, customer records, images, voice conversations, browsing behavior and other forms of personal data. The more useful the system becomes, the more tempting it can be to give it more context.
The Privacy Rule Most AI Users Ignore
Do not evaluate an AI service only by asking whether it is “private.” Check what data you send, whether it is retained, whether it can be used for model training, who can access it, and whether your account settings change those defaults.
Why More AI Exposure Doesn't Automatically Solve the Trust Problem
There is an easy assumption that people become comfortable with new technology through repeated exposure.
The Gallup results suggest something more nuanced.
Frequent users are less worried and substantially more trusting of AI accuracy than people who rarely use it. But the majority of daily users in the United States still report worry.
That means exposure may improve understanding without eliminating uncertainty.
And that makes sense when the technology itself is changing faster than users can build stable expectations around it.
The Real AI Trust Gap Is About Control
People do not necessarily need AI to be perfect before they will use it.
They need to understand what it is doing and where they remain in control.
That becomes particularly important as AI shifts from answering questions to taking actions. An AI that drafts an email is one thing; an agent that can send the email, access accounts or complete transactions carries a different level of responsibility.
Microsoft's 2026 responsible-AI reporting describes this broader shift toward agentic systems, where access, permissions, memory and actions become increasingly important parts of risk management.
| AI Use | Typical Risk | Useful Human Control |
|---|---|---|
| Brainstorming | Low factual reliability | Review before using |
| Research | Incorrect or outdated information | Verify important sources |
| Work automation | Wrong actions or data exposure | Limit permissions and require approval |
| High-stakes decisions | Material consequences from errors | Keep qualified human review in the loop |
What Generic AI Coverage Often Misses
1. Worry is not the same as rejection
A person can be excited about what AI can do personally and worried about what widespread deployment means for jobs, privacy or society.
2. Daily use does appear to correlate with greater trust
Gallup's figures show substantially higher trust in AI accuracy among frequent users, although the survey cannot establish which direction the relationship runs.
3. The United States has a distinctive sentiment profile
The U.S. combines high AI use with unusually high levels of worry compared with the other countries currently included in Gallup's 2026 results.
4. Privacy is not a future problem
Pew's finding that 71% expect AI to make personal information less secure shows that data protection is already central to public perception.
5. People are separating usefulness from trust
This may be the most important behavioral shift. Users do not have to believe AI is always correct to find it valuable.
A Useful AI Conversation Is Happening on X
Warning is good - scaring is less good. AI communication needs clarity, context and something people can act on.
— Jason Averbook (@jasonaverbook) March 31, 2026
The post above offers a useful framing for the current AI conversation: concern can be informative without becoming the only story people hear.
Amazon: Simple Privacy Hardware for AI Users
Webcam Privacy Cover
A simple physical webcam cover can reduce accidental camera exposure on laptops and external displays. It does not replace good account security or software privacy controls, but it is an inexpensive physical safeguard.
Check Webcam Privacy Covers on Amazon →Yubico Security Key
A hardware security key can add phishing-resistant authentication to compatible accounts. It is particularly relevant for AI users whose accounts may contain private conversations, business files or connected services.
Check Security Keys on Amazon →So, Are Americans Turning Against AI?
The available evidence does not support such a simple conclusion.
Americans are using AI more. Pew found that 36% of U.S. adults interact with AI at least several times a day, while roughly half report having used AI chatbots.
At the same time, concern remains high. Gallup finds 74% of AI-aware Americans worried about AI, while Pew finds 52% more concerned than excited about increased AI use in daily life.
Those numbers can coexist because adoption, usefulness, optimism, trust and worry are different dimensions.
That is the real story.
What AI Companies Need to Understand
The public does not appear to be demanding that AI disappear.
What the research shows is a demand for something more complicated: useful technology that people can understand, verify and control.
For developers, that means better explanations, clearer permissions, stronger privacy defaults and obvious ways to correct mistakes.
For users, it means learning where AI is reliable, where it is uncertain and when human judgment should remain decisive.
The Bigger Picture
AI has crossed an important threshold.
It is no longer something most people encounter only in technology demonstrations. It is appearing in search, office software, phones, smart devices, customer service, education and everyday personal tasks.
That makes public sentiment more consequential.
The question is no longer whether people have heard of AI.
The question is whether they trust themselves, their employers and the companies building these systems to use it responsibly.
And right now, the answer is clearly mixed.
Americans are using AI because it is useful. They are worried because they understand that useful technology can still create real risks.
Is the 2026 AI Bubble Actually Real?
Beyond public anxiety over jobs and data privacy, Wall Street and the Federal Reserve are raising alarms over massive infrastructure capital expenditure and depreciating GPU clusters. Read our complete 2026 AI Bubble investigation to examine hyperscaler spending, Federal Reserve warnings, and whether current valuations can hold.
Read the AI Bubble Analysis →Sources and further reading:
Gallup — AI Optimism Globally Widespread Despite Uneven Use
TechCrunch — Even Americans Who Use AI Every Day Are Worried About It
Pew Research Center — Americans and AI 2026
Pew Research Center — Young Adults in the U.S. Are Increasingly Wary of AI
Pew Research Center — Americans Have Grown More Concerned About AI Over Time
Reuters/Ipsos — Three Out of Four Americans Say AI Firms Are Not Doing Enough
Frequently Asked Questions About AI Trust in America
What percentage of Americans who use AI daily are worried about it?
Gallup found that 68% of U.S. adults who use AI daily say AI makes them feel worried. The share is higher among less frequent users.
How many Americans are more concerned than excited about AI?
Pew Research Center found that 52% of U.S. adults were more concerned than excited about the increasing use of AI in daily life in its June 2026 survey.
Do Americans who use AI trust its answers more?
Gallup found a strong relationship between frequency of AI use and trust in AI accuracy. In the U.S., 45% of daily users said they trust AI a lot or completely to provide accurate information, compared with 8% of people who have never used AI.
Why are Americans worried about AI?
Major concerns identified in recent research include potential job losses, privacy and data security, the accuracy of AI information, and broader effects on society and the economy.
Are Americans rejecting AI?
The research does not show a simple rejection of AI. Americans are increasingly using AI while also expressing concerns about its effects, meaning adoption and trust are developing at the same time.
No comments:
Post a Comment