Just Capital released the second edition of its monthly National AI Monitor, which asks Americans how they are experiencing AI in their daily lives and what drives positive and negative experiences. 

The results show that despite the heightened national dialogue around AI safety, Americans are positive about their experiences with AI on a day-to-day basis, with productivity at work cited as a primary driver. Worries about environmental impacts, the trustworthiness of models, and threats to employment weigh heavily on their outlook for the future, especially among women, lower-income Americans, and older adults.

“There’s an assumption that Americans are against AI, but the people we’re hearing from continue to tell a more hopeful story,” said Just Capital CEO Martin Whittaker. “Americans’ optimism may be one of the most underappreciated signals in the AI conversation today. Leaders who listen closely have a real chance to build public trust.”

September National AI Monitor results: 

• Overall: +37, up from +35 in August
• Work: +44, down from +48 in August 
• Customer experience: +25, up from +21 in August
• Community engagement: +12, up from +7 in August
• Investments: +24, up from +21 in August
• Environmental stewardship: -12, up from -17 in August 
• Over the next few years: +13, down from +15 in August 

Scores are calculated as the percentage of positive responses minus the percentage of negative responses, producing a range from -100 to +100. Changes between September and August were not statistically significant. 

Drivers of positive experience include: 

Productivity at work

“AI takes care of the structured and logical work; this allows me time for bringing human expertise to bear.” –  Male, 59, middle income, very positive respondent on AI at work

“I can do 10x the work I used to do, and can more confidently work in areas I dont have as much experience in.” – Male, 39, upper income, very positive respondent on AI at work

Speed, convenience, and personalization

“As a customer, I think companies’ use of AI makes products and services more convenient and personalised. It helps me get faster customer support, better recommendations, and quicker answers to my questions. At the same time, I am concerned about privacy and want companies to use my personal information responsibly.” – Female, 42, middle income, very positive respondent on AI as a customer

Exposure as an investor

“My 401k is doing pretty well and AI is driving the market.” – Male, 34, lower income, very positive about AI as an investor 

Drivers of negative experience include:

Environmental impacts including data centers, water use, electricity use, and grid strain

“Water waste, water pollution, energy waste, the drain on resources of communities too poor to fight back against them.” – Male, 50, upper income, very negative respondent on AI and the environment

Distrust

“I do not trust AI or the makers. I don’t think it can be controlled.” – Female, 65, lower income, very negative respondent on AI overall

Job loss 

“It’s very sad to see that AI is quite literally taking people’s jobs and copying people’s content and reusing their face and videos without their knowledge of consent.” – Female, 19, lower income, very negative respondent on AI at work

Americans are still positive about the impact of AI in the future, although optimism wanes markedly compared to sentiment today. 

Attitudes vary considerably by respondent demographics: 

• Women: 35% of women expected AI to improve their lives, compared with 54% of men. Women were slightly more likely to expect harm (36%) than benefit. Men were also more optimistic than women in August.

• Race: White Americans primarily expect a positive effect (42%) versus a negative one (36%). Black Americans, by comparison, are even more optimistic with 51% positive and 17% negative. White Americans were also less optimistic than Black Americans in August.

• Lower-income: 34% of those earning under $50,000 expected AI to improve their lives, compared with 44% to 48% in higher brackets. 23% of lower-income Americans were not sure. Optimism rose with income in August as well.

• Older adults and Gen Z: About half of Americans aged 25 to 54 expected AI to help them in the coming years, while both younger adults and those 55 and older were notably less optimistic. Among those 55 and older, the split between positivity and negativity was almost exactly even, which produced the lowest net score of any age bracket. 

• Democrats: Democrats were about evenly divided between expecting a positive (40%) and negative effect (38%). Republicans were more likely to expect a positive effect (51%) over a negative effect (29%), as were independents (43% vs 30%).

Methodology

The survey was conducted online within the United States by The Harris Poll on behalf of Just Capital from September 30 – October 2, 2026 among 1,010 U.S. adults ages 18 and older. The sampling precision of Harris online polls is measured by using a Bayesian credible interval. For this study, the sample data is accurate to within +/- 3.6 percentage points using a 95% confidence level. This credible interval will be wider among subsets of the surveyed population of interest.

Respondents were asked seven questions: one on AI’s overall effect on the respondent to date, one for each of the five stakeholder groups in Just Capital’s framework (workers, customers, communities, the environment, and shareholders), and one forward-looking question on how optimistic respondents are about where AI is headed and how it may affect them. All questions use a five-point scale ranging from “very negative” to “very positive,” with “I don’t know” offered as an additional, non-forced option so that uncertainty is not recorded as a neutral rating. Every question except the final one is followed by an open-ended “why,” which is synthesized to identify what drives responses and to serve as a diagnostic on how effectively companies are deploying AI. Scores are calculated as the percentage of positive responses (“somewhat positive” and “very positive”) minus the percentage of negative responses (“somewhat negative” and “very negative”), producing a range from -100 to +100. “I don’t know” responses are excluded from the calculation rather than counted as neutral.

For complete survey methodology, including weighting variables and subgroup sample sizes, please contact Just Capital.