How People Are Actually Using AI

A Surprising Shift from Productivity to Personal Purpose

As someone who builds AI training programs and automation tools for the workplace, I’ve always thought of AI in terms of efficiency, performance, and business outcomes. So when I read the Harvard Business Review
“Generative AI: How People Are Really Using Gen AI in 2025” by Mark Zeo Sanders, it really caught me off guard.

Not just because it was well-researched—based on actual user behavior from platforms like Reddit—but because the top three use cases people are turning to AI for in 2025 aren’t what you might expect.

• They’re not about productivity hacks.
• Not about automation.
• Not even about content creation.

They’re about being human.

A Shift at the Top: What People Are Really Using AI For

According to the report, Sanders and his team compiled the top 100 real-world applications of generative AI, and what they found paints a very different picture from the standard business use cases I am used to seeing.

Here are the top three:

1. Therapy and Companionship

“Large language models are accessible to everyone… they can help. Data privacy is less of a concern when survival is the morning agenda.” – A user in South Africa

In regions where access to mental health care is limited, AI is filling a crucial role. Not just as a chatbot or productivity tool, but as a non-judgmental source of emotional support and structured therapy. It’s available 24/7, doesn’t need appointments, and provides comfort without stigma. Some studies even suggest that AI-generated responses are approaching the quality of human therapists in specific contexts.

2. Organizing My Life

“AI is becoming like a personal assistant at work.” – Jared Spataro, Microsoft

This category is all about structure. From creating detailed cleaning schedules to habit tracking and goal setting, people are turning to AI for mental clarity and planning. Whether it’s through ChatGPT or Microsoft Copilot, AI is being used to reduce cognitive load and manage the practical side of daily life.

3. Finding Purpose

Perhaps the most unexpected—and revealing—category. Users are engaging AI to explore their values, set life goals, and get help reframing personal obstacles. It’s like having a personal coach or reflective sounding board always at hand. For many, this means using AI not just to get things done, but to become who they want to be.

These three uses alone indicate a major shift in how people are relating to AI. From functional tool to emotional partner. From data assistant to life coach.

Beyond the Top 3: Surprising Use Cases

The rest of the list continues to surprise. Some of the most fascinating examples that Sanders includes are:

• Enhanced Learning (#4): Using AI as a personalized tutor—explaining concepts glossed over in online courses.
• Healthier Living (#10): AI-generated recipes tailored to specific macro goals like protein intake or carb counts.
• Travel Planning (#24): Fully personalized itineraries, including hidden local gems.
• Disputing a Fine (#83): Drafting a successful appeal letter to overturn a parking ticket.
• Interacting with the Digital Deceased (#33): AI is being used to simulate conversations with lost loved ones, pushing the boundaries of memory and grief.

Each use case reflects an expanding trust and comfort with AI in personal, intimate, even emotionally complex areas of life.

From Business Tool to Human Tool

What really stayed with me after reading this report is how our perception of AI is evolving.

In my work, we focus on helping teams become more productive with AI. But this report highlights a parallel trend that’s just as important: AI as a mirror, a helper, a quiet partner in navigating life.

It’s not just about getting more done.

It’s about being more human.

Yes, business use cases like AI for tax prep, document creation, and workflow automation are still growing—and I believe they’ll continue to define a huge part of the AI economy.

But what people want from AI today is often far more personal, reflective, and emotional than we may have predicted.

Friction, Concerns, and the Road Ahead

The report doesn’t shy away from the tensions. It covers:

• Concerns around over-reliance and loss of critical thinking
• The ongoing privacy debate (some want more personalization, others want less data collection)
• The rise in sophisticated prompting as users become more savvy
• A growing desire for more agentic AI — tools that can take action for you, not just suggest things

It’s clear that as AI becomes more integrated into everyday life, the expectations—and the ethical questions—are only going to grow.

So, What do People Really Want from AI

This report gave me a new lens on AI. I wrote this blog because, as someone who’s been focused on AI for productivity, this shift toward AI for purpose felt like a wake-up call. People aren’t just using AI to write faster or answer emails. They’re using it to cope, to reflect, to grow.

And in the full Top 100 list below from Sanders, the breadth of these use cases shows a future where AI isn’t just embedded in the workplace—but in the fabric of how we live.

What does this mean for how we design AI tools, train users, or frame the next generation of AI services?

It’s a question I’ll be sitting with.

Maybe you will too.

The Full list of How People are using Gen AI from by Mark Zeo Sanders.

Source:
How People Are Really Using Gen AI in 2025” by Mark Zeo Sanders. Based on user experiences gathered from online communities including Reddit.

Insights from the podcast discussion of “Generative AI:” by Google’s AI Notebook LM