--- title: "New Research Reveals the Surprising Ways People Actually…" canonical: "https://www.metaintro.com/blog/how-people-really-use-ai-at-work-2026" language: "en" author: "drashtigarach" published: "2026-06-05T15:57:53.000Z" modified: "2026-06-05T15:57:53.000Z" --- [Back to Blog](/blog) [News](/blog/tag/news)[AI](/blog/tag/ai)[Technology](/blog/tag/technology)[Career](/blog/tag/career) # New Research Reveals the Surprising Ways People Actually Use AI at Work New HBR research reveals therapy, life-planning and purpose now top how people really use AI at work in 2026. See what the data means for your career. [![Drashti Garach](https://cdn.metaintro.com/rs:fill:40:40/q:72/plain/images/5719d740-e510-42bc-8017-e040d145f35f_1766029465094.png)Drashti Garach @DrashtiGarach](/blog/author/drashtigarach) [June 5, 2026](/blog/archive/2026/06)12 min read ![New Research Reveals the Surprising Ways People Actually Use AI at Work](https://cdn.metaintro.com/rs:fill:1200:675/q:78/plain/images/kai.OusDYmPl.png) [https://x.com/intent/tweet?text=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026](https://x.com/intent/tweet?text=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026)[http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026](http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026)[https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026&title=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026&title=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work)[mailto:?subject=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026](mailto:?subject=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026) A sweeping new study published in [Harvard Business Review](https://hbr.org/2026/06/how-people-are-really-using-ai-in-2026) confirms what many workers already suspect, the way people actually use AI on the job in 2026 looks very different from the polished demos vendors love to show. The research, led by [filtered.com](https://filtered.com/) co-founder Marc Zao-Sanders as the latest edition of his ongoing [AI in the Wild](https://marczaosanders.com/wild/) project, finds that personal and emotional support has overtaken the technical, code-and-spreadsheet tasks that once dominated. Companionship, life organization, and the search for purpose now sit near the top of the list, while raw productivity gains in the office remain, in the authors' own framing, more marginal than game-changing. At [Metaintro](https://www.metaintro.com), we read this so you can make smarter career moves, and the headline is simple. The tools your coworkers reach for are often softer and more human than the hype suggests. That gap between the demo and the desk matters for anyone planning a career around AI. If you assume everyone is shipping autonomous agents and rewriting their job overnight, you may panic. The real picture is messier, slower, and far more usable. Below we break down what the research actually found, where the numbers come from, and what it means for the way you work, apply, and skill up this year. ## What Did the New AI Research Actually Find? The most striking finding is a shift in motivation. In 2024, the top reason people gave for using generative AI was simply to generate ideas, a classic brainstorming use. By [2025](https://hbr.org/2025/04/how-people-are-really-using-gen-ai-in-2025), therapy and companionship had climbed to the number one spot, and the 2026 edition shows that personal and professional support now accounts for roughly 31 percent of all the use cases the team catalogued. That single category, covering everything from talking through a hard week to rehearsing a tough conversation, has become the center of gravity for everyday AI use. Two other entries near the top of the list tell the same story. "Organizing my life" and "Finding purpose" both rank among the most common reasons people open a chatbot, sitting alongside therapy as expressions of self-actualization rather than spreadsheet labor. People are using these tools to process emotions, reflect on personal beliefs, and prepare for the parts of life that have nothing to do with quarterly targets. The authors describe a clear migration from technical applications toward human connection and personal growth. None of this means the work tasks have vanished. Coding help, writing assistance, and search-style lookups still fill out the rankings. But the research frames the office gains as incremental so far, a useful corrective for anyone who has been told AI will double their output by Friday. The more durable change is how casually people now fold these tools into their inner lives. The scale behind these habits is enormous, which is part of why the shift matters. The article notes that [ChatGPT](https://openai.com/) now reaches roughly 900 million weekly active users and [Google Gemini](https://gemini.google.com/) around 750 million monthly users, with OpenAI valued at about 852 billion dollars in its latest funding round. When tools at that scale tilt toward personal and emotional use, the workplace inherits the pattern by default. Your colleagues are not waiting for an enterprise rollout to form their habits. They are bringing the muscle memory of nights-and-weekends chatbot use straight into Monday morning. ## Where Does the Data Come From and Can You Trust It? Methodology matters when a study makes claims this bold, so it is worth understanding how this one was built. Rather than running a single survey, Zao-Sanders and his collaborators curate real-world discourse, mining online forums where people describe, in their own words, what they actually do with AI. [Reddit](https://www.reddit.com/) and similar communities like [Quora](https://www.quora.com/) have consistently yielded the richest detail, supplemented by articles that document specific, named applications of the technology. The approach captures behavior people rarely admit in a formal poll. The 2026 edition widens the lens considerably. The team expanded beyond forums to include corporate surveys, its own web-based collection form, and a partnership with the learning company [Mindstone](https://www.mindstone.com/), and it gathered evidence over a full twelve months rather than the roughly six-week window used in earlier editions. That longer runway and broader sourcing make the new rankings more robust than the viral 2025 version, which still racked up hundreds of thousands of readers. It is qualitative work at heart, so treat the ranks as directional rather than census-grade, but the consistency across three years gives the trend real weight. Outside research points the same direction. A [KFF](https://www.kff.org/) poll found that 16 percent of U.S. adults turned to AI tools or chatbots for their mental health in the past year, with higher rates among younger people, and a [Pew Research Center](https://www.pewresearch.org/) survey found that about two-thirds of teens interact with chatbots, with 28 percent messaging daily. The pattern is significant enough that, as [NPR](https://www.npr.org/2026/04/10/nx-s1-5780796/artificial-intelligence-ai-mental-health-chatgpt-jama-psychiatry) has reported, some clinicians now ask patients how they use AI. When independent pollsters and a forum-based study converge, the underlying behavior is hard to dismiss as noise. ## Why Are People Using AI for Emotional Support at Work? The pull toward emotional and personal use is not hard to explain once you look at the alternatives. Large language models are available at 2 a.m., they are anonymous, they do not judge, and they carry none of the social cost that comes with admitting to a colleague that you are overwhelmed. For a worker between therapy appointments or stuck on a stressful decision, a chatbot becomes a private space to untangle thoughts before facing the room. One user quoted in coverage of the research described using AI to piece together "fractured" thoughts during moments of emotional upheaval. That accessibility is also why the trend bleeds into the workday. The line between personal and professional support is thin. The same tool that helps someone draft a difficult email to a manager is the one they use to calm their nerves before sending it. As we explored in our piece on whether [AI is deskilling healthcare workers](https://www.metaintro.com/blog/is-ai-deskilling-healthcare-workers-nurses-clinicians), the convenience of always-on AI can quietly reshape habits in ways that are easy to adopt and hard to notice. The researchers are careful not to celebrate this uncritically. They flag a rising anxiety, which the latest edition labels "thinkslop," that people may be surrendering their own cognitive responsibilities to the machine, leaning on it for judgment and reflection they used to do themselves. There is a parallel worry about over-reliance for emotional support. The lesson for workers is not to avoid these tools but to stay aware of where the help ends and your own thinking should begin. It is worth being precise about what "therapy" means in this context, because the word can mislead. The study is not claiming that millions of people have replaced licensed counselors with a chatbot. It is documenting that people use AI for low-stakes emotional processing, venting after a bad meeting, drafting what they want to say to a difficult coworker, or making sense of a stressful stretch before they talk to an actual professional. That distinction matters for your career, because the same reflective skill that makes someone good at this private use, asking better questions and sitting with an answer, is exactly the skill that makes them better at managing ambiguity at work. ## What Does This Mean for Your Career? For job seekers and employees, the practical takeaway is reassuring and a little contrarian. The fear that AI has already automated everyone's job into oblivion does not match how people actually use it. The dominant uses are augmentation and support, not wholesale replacement, and the office productivity story is still a work in progress. That gives you room to build real, lasting skills rather than chasing every new agent demo. Here is how to put the findings to work: - **Lead with judgment, not just prompts.** If the research shows people leaning on AI for ideas and emotional processing, the scarce skill becomes deciding what is actually good. Employers increasingly value people who can edit, verify, and improve what AI produces. - **Name your AI fluency honestly.** Knowing the everyday workflows, drafting, summarizing, organizing, planning, is more credible in interviews than vague claims about building autonomous systems. As we noted in our look at why [AI alone isn't blocking young workers from jobs](https://www.metaintro.com/blog/ai-isnt-why-young-workers-cant-find-jobs-study), the labor market rewards demonstrated capability over buzzwords. - **Guard against thinkslop.** Use AI to accelerate your thinking, then pressure-test its output. The professionals who stand out keep their own reasoning sharp, a point that runs through our coverage of how [small businesses are adopting AI](https://www.metaintro.com/blog/small-businesses-ai-employees-instead-of-hiring) without abandoning human oversight. The broader signal is that AI is becoming a normal, daily companion rather than a sci-fi replacement, and the workers who thrive will be the ones who use it deliberately. For a fuller view of where hiring is heading this year, our [state of the workforce report](https://www.metaintro.com/blog/state-of-the-workforce-may-2026) tracks the trends shaping demand. ## How Should Employers Read These Findings? Companies that want a return on their AI spending should pay attention to the gap between expectation and reality. The research suggests that most business value so far is marginal, a sign that simply buying licenses does not transform output. The organizations seeing gains tend to pair the tools with training and clear workflows, helping employees move past the brainstorm-and-stop pattern that shows up in the data. There is also a duty-of-care dimension. If a meaningful share of usage is emotional support, leaders should think about boundaries, privacy, and the risk of staff substituting a chatbot for real human and professional help. The same dynamics we examined in our reporting on [Microsoft's wearable AI badge for office workers](https://www.metaintro.com/blog/microsoft-wearable-ai-badge-office-workers) apply here. Adoption moves faster than policy, and the smartest employers close that gap on purpose rather than by accident. For workers, the implication is opportunity. Professionals who can help a team adopt AI thoughtfully, designing good prompts, setting guardrails, and coaching colleagues, are filling a gap that most companies have not staffed yet. That is a career lane worth watching as the technology settles into ordinary work. You do not need to be an engineer to claim it. Many of the most valuable contributors are people who simply understand how their team really works and can translate the technology into habits that stick. ## What Should You Do Differently This Week? Start by auditing your own AI habits with clear eyes. Notice which tasks you genuinely offload and which ones you should still own, and treat the tool as a draft partner rather than a final authority. The research is a reminder that the most common uses are quiet and personal, so there is no shame in using AI to organize a chaotic week or think through a hard decision, as long as you stay in the driver's seat. Then translate that awareness into how you present yourself. In a market where everyone claims to "use AI," specifics win. Describe the real workflows you rely on and the judgment you bring to them. The candidates who pair fluency with discernment are the ones who will stand out as the dust settles on the hype cycle. Finally, keep the longer arc in view. Three editions of this research now tell a steady story, ideas in 2024, companionship and self-actualization by 2026, and incremental rather than explosive office gains throughout. That trajectory rewards patience over panic. Skills compound, and the workers who treat AI as a durable daily companion they have learned to direct will be far better positioned than those who chase each viral demo and burn out. The hype cycle moves fast, but real capability is built slowly, and this study is a strong argument for betting on the latter. ## People Also Asked ### Q: What is the number one way people use AI in 2026? A: According to the HBR research led by Marc Zao-Sanders, personal and emotional support, including therapy and companionship, has become the leading category, making up roughly 31 percent of catalogued use cases and overtaking the technical office tasks that dominated in 2024. ### Q: Does this research mean AI is replacing therapists? A: No. The study documents low-stakes emotional processing, like venting after a hard day or rehearsing a difficult conversation, not the replacement of licensed care. Many people use AI to organize their thoughts between real therapy sessions, and the authors flag over-reliance as a genuine risk. ### Q: How was the AI use case data actually collected? A: The team curated real-world discourse from online forums such as Reddit, plus articles describing specific applications, and the 2026 edition added corporate surveys, a web collection form, and a Mindstone partnership over a full twelve months. It is qualitative, so the rankings are directional rather than a precise census. Future-proof your career with [Metaintro](https://www.metaintro.com), AI-powered tools that help you adapt and thrive. [Sign up](https://www.metaintro.com/signup) to stay ahead. ### Share this article [https://x.com/intent/tweet?text=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026](https://x.com/intent/tweet?text=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026)[http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026](http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026)[https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026&title=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026&title=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work)[mailto:?subject=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026](mailto:?subject=New%20Research%20Reveals%20the%20Surprising%20Ways%20People%20Actually%20Use%20AI%20at%20Work&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fhow-people-really-use-ai-at-work-2026) For job seekers ## The right role, found for you. Built for people who’d rather be good at their job than good at job hunting. [Get Started Free](/signup)[Search matching jobs](/jobs/search) - MatchCompare live roles against your current evidence. - PositionTurn proof projects into role-specific applications. - ImproveUse market feedback to keep the skill plan current. [Return to navigation](#main-navigation)