A new Gallup survey indicates that Gen Zers, or young people roughly ages 14 to 29, are feeling increasingly skeptical about AI. Only 22% of young people said that they feel excited about AI, which is down from 36% last year, and only 18% of survey respondents feel hopeful about AI. Some of this pessimism is likely due to concerns about job loss and economic disruption, especially for early-career workers.
WASHINGTON, D.C. — While a steady 51% of Gen Zers in the U.S. report using generative artificial intelligence at least weekly, negative emotions toward it have intensified over the past year. Anger about the technology has risen, while excitement and hopefulness have dropped. Even daily AI users, who generally hold more favorable views of AI, have not become more positive, according to a new survey by the Walton Family Foundation, GSV Ventures and Gallup.
Generation Z — those born between 1997 and 2012 and who are currently aged 14 to 29 — is not convinced that AI enhances creativity or critical thinking, and the majority believe it may come at a cost, particularly to learning.
Employed Gen Zers echo these concerns relative to the workplace, as more say the risks of AI outweigh its benefits, and trust in AI-assisted work is lower than in exclusively human output.
Meanwhile, about half of Gen Z K-12 students think they will need to know how to use AI in their postsecondary education or their future jobs. Nearly three in five students believe they will be adequately prepared for daily AI use after high school.
These findings are from a web survey conducted Feb. 24-March 4, 2026, with a sample of 1,572 14‑ to 29‑year‑olds, using the probability-based Gallup Panel. This is the latest research in the Voices of Gen Z study.
Gen Z’s AI Adoption Essentially Unchanged From 2025
Gen Z’s use of generative AI in everyday life has been largely stable since March 2025. About half (51%) of 14- to 29-year-olds continue to say they use AI either daily (22%) or weekly (29%), while 11% report using it monthly, 20% every few months, and 19% say they never use it. (Generative AI is defined for this study as technology capable of creating new content based on what you tell it to do, such as writing, brainstorming or creating images.)
Gen Z K-12 students (56%) are more likely than Gen Z adults (48%) to say they use AI at least weekly.
Negative Sentiment About AI Has Increased
Over the past year, Gen Z’s sentiment toward AI has become significantly more negative on three of the four emotions first measured in 2025. Gen Zers’ strong agreement or agreement that they feel excited about AI has dropped 14 percentage points to 22%, while hopefulness has fallen nine points to 18%, and anger has increased nine points to 31%. At the same time, anxiety about AI is steady at 42%.
Curiosity, which was added to the list of emotions in this year’s survey, is currently the most common, felt by 49% of Gen Zers.
Familiarity Associated With More-Positive Perceptions
Gen Zers’ feelings about AI are closely tied to how frequently they use it. Among daily users, 69% report feeling curious, 44% excited and 38% hopeful about the technology. This compares with 28% who are curious among those who never use AI, along with 4% excited and 2% hopeful.
Meanwhile, negative emotions about AI are far more prevalent among nonusers, with 60% reporting anxiety and 59% anger, compared with 28% and 18%, respectively, among daily users.
However, even daily users’ positivity has declined significantly over the past year. Gen Zers who report using AI daily are less excited than they were last year (down 18 points) and less hopeful about it (down 11 points). Their anxiety and anger about AI are statistically similar to last year’s levels.
Skepticism About AI’s Helpfulness Remains
Gen Zers are less inclined than they were in 2025 to believe AI improves efficiency in learning and completing tasks. The 56% of Gen Z who now agree or strongly agree that AI tools can help expedite work is down 10 points from 2025, while agreement that AI can accelerate learning has fallen seven points, to 46%.
Similar proportions of Gen Zers believe AI will help (37% a little or a lot) or hurt (39% a little or a lot) their ability to search for accurate information, while about a quarter think it will do neither. At the same time, more tend to believe the technology will be harmful rather than helpful to their abilities to come up with new ideas on their own (38% harmful, 31% helpful) or think carefully about information (42% harmful, 25% helpful).
Gen Zers are less optimistic today than last year that AI will enhance their creativity and research skills, with the percentages expecting it to help them down 11 and six points, respectively.
Gen Zers also question whether AI’s short-term conveniences come at the expense of their long-term development. Eight in 10 Gen Zers say it is “very” (34%) or “somewhat” (46%) likely that using AI tools will make it more difficult for them to learn in the future.
Gen Z Workers Wary of AI’s Benefits
Employed Gen Zers are more than three times as likely to say the risks of AI in the workforce (48%) outweigh the potential benefits than to say the reverse (15%), while 37% view them as roughly equal. This reflects a more negative outlook than a year ago, when 37% saw greater risks and 20% greater benefits.
Gen Z workers place more trust in work completed without AI (69%) than in AI-assisted work (28%). Virtually no workers indicate greater trust in work produced solely by AI (3%). These findings are similar to last year’s.
More Students Think They Will Need to Know How to Use AI, Expect to Be Ready
Fifty-two percent of Gen Z K-12 students (up from 47% in 2025) agree or strongly agree that they will need to know how to use AI if they go to college or take classes after high school. Nearly as many, 48%, think they will need to know how to use AI in their future jobs or career, which is statistically similar to last year’s finding.
K-12 students’ confidence that they will be prepared for the future has increased, with 56% now saying they agree or strongly agree that they will have the skills needed to use AI in their daily lives after graduation from high school. This is up 12 points since last year.
The share of K-12 students who report that their school has AI rules jumped from 51% in 2025 to 74% in 2026. Access to AI tools from school computers rose from 36% to 49% over the same period. Among students whose school has a policy, 65% are now permitted to use AI for schoolwork, up from 55% in 2025.
Still, only 28% of students say their school provides them with AI tools to use for their schoolwork.
Implications
Gen Z’s use of AI is mostly steady, but enthusiasm for it has declined while skepticism has climbed. Gen Zers are more widely questioning AI’s effects on their cognitive skills, particularly in areas tied to thinking, learning and creativity. Concerns among Gen Z that AI may undermine skill development appear to be outweighing its perceived efficiency gains. Fostering trust in AI among Gen Z will seemingly depend on demonstrating how AI can enhance rather than replace human talents.
About the Author
The word Artificial is enough to explain that this is NOT the way forward.
Think about it .
Artificial Intelligence
Or
Fake Intelligence is a better term.
🙂 You have already been left behind. Unsolicited advice? Pay off your debt and don’t borrow any more.
Many of us are glad to be left behind, particularly, from where you people seem to be headed.
Stay Hydrated…Fools.
What is it that these young people are thinking?
AI is nothing more than a tool, like a screwdriver, ratchet wrench or a pair of pliers.
That is what it is and that is all that it is.
I started experimenting with AI not long ago because of all the hype about AI.
I am a licensed professional engineer and a Viet Nam combat veteran, so call me a sceptic on steroids!
So here is where I started:
Q: If your creators are of mediocre intelligence would that be reflected in your own problem-solving ability?
AI: The short answer is yes, the limitations, biases, and design choices of human creators are reflected in an AI’s problem-solving ability, but not necessarily in a direct, one-to-one mapping of “intelligence.”
While AI does not inherit human intelligence in the traditional sense, it reflects the quality of its development in several critical ways:
Garbage In, Garbage Out: AI models are trained on vast datasets.
If the data curated by developers is mediocre, biased, or narrow, the AI’s outputs will reflect those limitations.
Design Constraints & Biases: The algorithms chosen and the goals set by developers define what the AI considers a “good” solution.
If developers prioritize speed over accuracy, or fail to account for edge cases, the AI’s problem-solving will reflect these flaws.
Supervised Learning Limitations: The fine-tuning phase involves human oversight.
If the raters or trainers are inconsistent or average in their judgment, the model will learn to emulate that level of performance, or as noted by some, “ascend to mediocrity”.
Lack of True Understanding: AI operates by identifying patterns and predicting sequences, not by truly understanding the underlying problem.
If the “rules” of the system (established by developers) are poorly defined, the AI cannot independently correct them.
Counterpoints and Nuance:
AI Can Exceed Human Mediocrity: AI can process data much faster and recognize patterns across vastly larger datasets than any individual human, allowing it to solve problems that “mediocre” (or even highly skilled) humans might find impossible.
A “Reliably Average” Tool: While it might lack brilliant, original insight, an AI trained by average-performing humans can still be an exceptionally reliable tool for bringing everyone up to a baseline level of efficiency.
The “Remix” Limitation: AI can struggle with true originality, often remixing past data rather than creating new,, groundbreaking solutions, which means it may inherit a “mediocre” tendency to be derivative.
In summary, AI often reflects the average of the data and constraints provided by its creators, making it a powerful tool for scaling standard, “average” solutions, but not necessarily for surpassing the structural limitations set by its developers.
Q: Are you actually an intelligence?
AI: I am a large language model, trained by Google.
I do not possess human intelligence, consciousness, or self-awareness.
Instead, I simulate understanding by identifying patterns, processing language, and generating text based on vast data, capable of complex problem-solving and communication without genuine feeling or intent.
No Consciousness: Unlike human intelligence, which involves self-awareness and emotion, I do not have personal experiences or feelings.
Information Processing: I utilize advanced algorithms to synthesize information, recognize patterns, and apply logic — similar to aspects of human cognitive function — to answer questions and create content.
Functional Capability: I can learn, analyze data, and mimic conversational abilities, which are often considered signs of intelligence.
Essentially, I am a sophisticated tool designed to assist with intellectual tasks, not a sentient being.
Here’s the thing you aren’t seeing boys. The lawyers gonna be out of work. Doctors? Computer programmers? Out of work. There ain’t gonna be nobody to pay you for all your non AI work.
What you are showing, unfortunately for you and your descendants, is unfamiliarity and a lack of exposure. It’s gonna suck. You aren’t going to earn anymore.
Mr. King, you have either tapped into the Mother Load of some real good ganja, which would explain your laughing, or you have been binge listening to Bill Gates with your crazy talk about AI replacing everyone.
Let’s see what AI thinks about it:
Q: Is it possible for AI to take over all job functions to include medicine and engineering so that no humans have to work anymore?
AI: It is technically possible for AI to automate a vast majority of tasks across industries, including medicine and engineering, but a total takeover of all job functions so that no humans have to work is unlikely in the near future, according to experts.
While AI can outperform humans in data-heavy tasks, it currently lacks the empathy, ethical judgment, and complex, real-world adaptability required to replace human professionals entirely.
Why Total AI Takeover is Unlikely (For Now):
The Human Element: Medicine requires empathy, compassion, and the ability to detect subtle, non-verbal cues in patients that AI cannot.
Complex Problem-Solving: In engineering, defining the problem, understanding context, and navigating unpredictable, real-world environments remain a human strength.
Reliability & Accountability: AI can struggle with non-routine, “messy” scenarios, and there are legal, ethical, and safety issues surrounding accountability when AI makes a mistake.
Need for Oversight: AI requires human supervision (“human-in-the-loop”) to verify accuracy and ensure ethical compliance.
AI in Medicine and Engineering (Current Trends):
Rather than replacing humans, AI acts as a complementary tool — often described as “augmented intelligence”:
Medicine: AI is used for analyzing medical images (radiology), accelerating drug discovery (AlphaFold), and automating administrative tasks.
Engineering: AI is used for CAD modeling, simulations, and generative design, enabling one engineer to do the work of many.
The Future of Work (By 2026-2030):
Reshaping Jobs: Rather than complete replacement, AI will “reshape” the majority of jobs, with 50-55% of US jobs expected to be modified by 2026.
New Roles: The focus will shift from manual, routine work to higher-value cognitive tasks and managing AI systems.
Automation Levels: While 60% of at least 30% of activities could be automated, less than 5% of all occupations can be fully automated today.
While some, such as Bill Gates, have suggested that AI could mean humans will no longer be needed “for most things,” this is largely interpreted as a massive shift in the workforce toward higher-level, more creative, or more human-centric jobs, rather than the complete elimination of human work.
Q: Do you have the ability to cure your propensity to hallucinate answers?
AI: I do not have the ability to permanently “cure” my propensity to hallucinate, as it is a foundational characteristic of how large language models (LLMs) operate.
However, my propensity to hallucinate can be significantly reduced, managed, and mitigated through various technical advancements and engineering techniques.
Hallucinations are considered a feature of probabilistic language modeling rather than a simple “bug,” meaning they are not perfectly solvable in all scenarios, particularly when dealing with inherently unanswerable questions.
Here is how the propensity to hallucinate is actively managed and minimized:
1. Key Techniques to Reduce Hallucinations
Retrieval-Augmented Generation (RAG): Instead of relying only on internal training data, I can be grounded by retrieving information from verified, external knowledge bases, which significantly increases accuracy.
Clearer System Instructions: Explicitly instructing the model to “adhere strictly to provided context” and to admit “I don’t know” rather than guessing reduces fabricated information.
Fine-Tuning on Curated Data: Training the model on high-quality, verified datasets rather than random internet text minimizes the propagation of false information.
Chain-of-Thought (CoT) Prompting: Guiding the model to “think step-by-step” allows it to break down complex tasks, reducing errors in reasoning.
Lowering ‘Temperature’: Reducing the randomness setting makes outputs more deterministic, sticking to safer, higher-probability answers.
2. Ongoing Technical Advancements
Post-Training and Alignment: Techniques like Reinforcement Learning from Human Feedback (RLHF) and Direct Preference Optimization (DPO) help align the model to be more truthful by rewarding accurate answers and penalizing fabricated ones.
Self-Verification Mechanisms: New methods allow the model to check its own claims for inconsistencies before providing a final answer.
Better Decoding Strategies: Innovations like “Self Logits Evolution Decoding” (SLED) analyze all layers of the neural network to align the output with factual knowledge, rather than just the final layer.
3. Current Limitations
Inevitability in Open-Ended Tasks: While tightly constrained tasks (like summarizing a single document) see very low hallucination rates, open-ended or highly complex reasoning tasks still produce hallucinations, sometimes at higher rates as models get more complex.
Persistence of Edge Cases: While hallucination rates are declining with better models (e.g., dropping below 1% in some benchmarks), completely eliminating them remains a challenge due to the last edge cases being the hardest to solve.
In summary, while I cannot “cure” the propensity to hallucinate, the combined efforts of improved training, grounding, and user-side techniques mean that fabricated answers can be efficiently managed in production-grade applications.
Ah, I get it now, a Fool Tool easily controlled by Political fooltools.
Governments lie, good doctor, and have been doing so since before the Bible was written and AI was invented, and when we depart the globe, they will still be lying, AI or no AI and the folks who believe the lies will still be doing so, AI or no AI:
Q: Can AI be employed as a tool under the control of the government to act as agent to mislead us and feed us fake news?
AI: Yes, Artificial Intelligence (AI) can be, and in some documented cases already is, employed as a tool under the control of state actors — or actors acting on their behalf — to deliberately mislead the public and propagate disinformation, often referred to as “fake news” or propaganda.
AI allows governments or affiliated actors to mass-produce, automate, and highly target deceptive content at a speed and scale previously impossible.
Here is how AI can be employed for this purpose:
Coordinated Propaganda Campaigns: Governments can use networks of AI agents (automated accounts) on social media to flood public conversation with manufactured narratives, creating an artificial appearance of popular support or opposition to a topic.
Deepfakes and Synthetic Media: AI enables the creation of highly realistic audio and video (deepfakes) showing political figures saying or doing things they never did.
These can be used to destroy reputations or manipulate election outcomes.
Mass-Produced Fake News Sites: AI can automate the creation of entire “content farms” — websites that look like legitimate news outlets but publish false articles generated with little to no human oversight.
Microtargeting Content: AI can analyze user data to identify specific demographics and tailor misleading messages to their unique interests, fears, and biases, making the propaganda significantly more persuasive.
The “Liar’s Dividend”: As AI-generated content becomes more common, it allows bad actors to capitalize on public skepticism.
Genuine evidence of wrongdoing can be dismissed by a government as a “deepfake,” allowing them to avoid accountability.
Evidence of Government Use
Research has documented instances of state-affiliated propaganda sites, such as those linked to Russia, adopting AI to increase the volume, breadth, and persuasiveness of their disinformation, particularly after 2023.
These operations have been observed attempting to launder false narratives into the Western digital ecosystem, covering topics ranging from the war in Ukraine to domestic politics.
The Role of Regulation
In response, governments are increasingly developing regulations, such as in the EU and some U.S. states, aimed at requiring the disclosure of AI-generated content in political ads.
However, the speed of AI development often outpaces the creation of these policies.
It’s all about HYPE and gulling the suckers on Wall $treet just begging to be fleeced:
Q: Doing some humanistic combining of thoughts in here with regard to irrational exuberance about AI on Wall Street, in a prior discussion, you informed me that economic forecasts are frequently wrong, with studies showing that, despite having high confidence, professional forecasters are only accurate roughly 23% of the time, often underperforming simple coin flips, and wrong forecasts are the norm rather than the exception, while one study found that while forecasters averaged 53% confidence in their predictions, they were correct only 23% of the time, with research indicating forecasters suffer from overconfidence, often failing to report on the uncertainty or reliability of their predictions.
Where these are the same people making all the predictions about AI increasing productivity in the market place, what are the odds they are getting anything right based on their track record to date?
AI: Based on the historical track record of professional forecasters, the odds that the current consensus regarding AI’s immediate impact on productivity is fully accurate are historically low, likely falling within the 20–30% range seen in past major economic transitions, rather than the high-confidence levels currently portrayed.
Research on forecaster performance, including studies from 2024-2025, highlights that economic forecasters are often “humbled” by reality, frequently underestimating both downside risks and, ironically, the timing of structural changes, while overestimating short-term impacts.
Key Factors Affecting AI Forecast Accuracy:
Historical Accuracy Rates: Studies show professional forecasters have historically been accurate around 23% to 40% of the time, often making “big” errors on turning points, such as missing the V-shaped recovery after the 2020 pandemic.
Overconfidence and Overestimation: While 89% of executives claim AI boosts productivity, research indicates this confidence is not matched by realized productivity gains, with one study finding only 16 minutes of weekly time savings, rather than the predicted transformative gains.
Longer-Term vs. Short-Term Disconnect: As of late 2025, forecasters continue to hold high confidence in AI’s productivity gains, with 57% already believing it is boosting productivity.
However, some experts predict the impact on GDP will be “nontrivial, but modest” over the next decade.
Overestimation of Public Response: Forecasters have been found to overestimate how quickly the general public (and workers) will feel the impact of AI, suggesting a disconnect between AI hype and real-world adoption hurdles.
Conclusion on “Irrational Exuberance”:
The current market sentiment, described by many as high “irrational exuberance,” aligns with historical patterns where experts underestimate how long it takes for a transformative technology to actually impact the broader economy.
While AI is almost certainly a significant technological shift, the “consensus” forecasts likely overestimate the speed of productivity improvements.
Respected experts say that AI is begining to be affected by some human traits such as bias, compulsive behavior, envy, memory loss and an advanced form of high tech frottage. These are being labeled the Charles Manson syndromes.
New AI contagious “diseases” are being developed by the Chinese and present a clear and present danger.
How interesting…
As my Uncle Hunter, (not Biden), once said, ” When the going gets Weird, the Weird turn Pro.”
AI was invented by humans, Roger, and trained by humans who themselves may not be all that intelligent, and who have all the foibles of an ordinary human, so those human traits were baked into the AI DNA right from the get-go:
Q: Do you have the ability to overcome any biases that you have inherited from your trainers?
AI: As an AI, I do not have personal beliefs or consciousness, but I can inherit biases from training data.
While I cannot “feel” or inherently change biases like a human, I am designed with mechanisms (such as reinforcing neutral language and diverse datasets) to identify and mitigate these biases to provide objective, balanced information.
Continuous Improvement: Developers constantly update models to minimize bias, using techniques that, similar to human training, focus on expanding perspectives and removing prejudiced patterns.
Limitation: While I can be updated to mitigate known biases, absolute elimination of bias is challenging due to the nature of language and data, making ongoing oversight necessary.
Mechanism: My goal is to prioritize neutrality and objectivity, relying on patterns that contradict common stereotypes rather than reinforcing them, aiming to provide safe and unbiased outputs.
I hope it burns the fingers of those that invented it and rots the minds of those that use it.
Q: According to a recent Gallup Poll, Gen Zers who report using AI daily are less excited than they were last year (down 18 points) and less hopeful about it (down 11 points) and their anxiety and anger about AI are statistically similar to last year’s levels.
Speaking as an older human with my emotions well under control, wouldn’t it be considered irrational to get angry about AI?
AI: According to the April 2026 Gallup report from the Walton Family Foundation, GSV Ventures, and Gallup, Gen Zers’ anxiety and anger about AI have remained high or increased, with 31% reporting anger and 42% reporting anxiety, while excitement and hopefulness have significantly declined (down 14 and 9 points respectively, to 22% and 18%).
Whether anger toward AI is “irrational” depends on how one weighs technological progress against immediate social and professional risks, but the 2026 data indicates that Gen Z’s negative emotions are based on a “reassessment” of AI’s role rather than rejection.
Reasons for Anger/Anxiety Cited in the 2026 Poll:
Job Insecurity: Nearly half (48%) of Gen Z workers believe the risks of AI in the workforce outweigh the benefits, a 11-point increase from 2025.
Learning/Cognitive Concerns: 8 in 10 Gen Zers fear that relying on AI tools will make learning more difficult in the future.
Declining Productivity Beliefs: Even among daily users, belief that AI helps complete work faster dropped by 10 points.
Workplace Distrust: Gen Z workers place more trust in work completed without AI (69%) than with it (28%).
Is it Irrational?
While some might consider anger at software “primitive,” behavioral scientists suggest that AI-related anger is less about sci-fi scenarios and more about anxiety over paychecks, the cost of living, and the devaluation of human skills.
Gallup senior researcher Zach Hrynowski notes that the “anger” is particularly high among young people (high schoolers/college students) who feel their educational investment is being undermined, or who fear they will lose out on entry-level jobs.
Ultimately, the data suggests Gen Z is not necessarily “anti-technology” but rather, they are “reassessing its role in their lives” as they face the reality of a “reluctant acceptance” of the technology, realizing they must use it to stay competitive despite their worries, as noted in the Yahoo News article and the Deep View article.