10 areas where AI will transform the way you work

Today I want to share ten key areas that anyone who wants to get the most out of artificial intelligence should focus on. You might think these are only for advanced users, but the opposite is true. The difference comes down to just a few extra hours a week spent experimenting and playing with AI. And the gap between people who know how to work with AI and those who don’t is growing fast.

1. Developing an AI mindset

This is the single most essential element of successful AI work.

We often expect AI to handle everything from A to Z: we throw in a presentation, say “redo it,” and wait for a new one. That’s not how it works, at least not in 2025. AI can help at every step of your work, but you need to learn how to communicate and engage with it. Sometimes you need to add an extra step, for instance, if it can’t handle a PDF directly, first extract the text, then feed it to AI.

Practical tip: Use AI to write prompts for AI. I have my own saved “prompt for writing prompts.” Once you start using something like it, you’ll find that almost everything can become part of that prompt.

A real example: I had an hour-long discussion with a colleague about the direction of our internal AI assistant. I took the transcript, had AI create a prompt to analyze our conversation, and extracted from it a description of what an AI mindset looks like. Then I asked how to measure it in job interviews. AI suggested a measurement approach, I wrote a brief for an AI agent, a colleague built a chatbot for measuring AI mindset, and now we’re integrating it into our product. Three prompts with feedback loops = a new product.

2. Choosing the right tools

I recommend working with all the major tools, even if your company only provides Copilot or one specific platform. If you want to stay relevant in the job market, you need experience with multiple tools. More importantly, each tool has a slightly different personality and skillset:

Claude – has its own personality, better grasp of complex requirements, more human-sounding texts, can build micro-apps
ChatGPT – browses the web, connects to a range of add-on tools, analyzes data, creates images
Google NotebookLM – analyzes large volumes of data (PDFs, YouTube videos) and cites sources
Perplexity – excellent for finding information and sourced results
Google Gemini – handles longer texts well
Microsoft Copilot – works seamlessly with Microsoft applications

Practical tip: Run “competitions” on complex projects. I give the same prompt to ChatGPT, Gemini, and Claude, and it’s fascinating to watch how differently each model works. It’s like having three extremely smart colleagues tackle the same problem.

Bonus: You can build your own AI tools. I discussed with AI how something could work, AI produced a description, I handed it to an app-building tool, and within 20–25 minutes I had a capable assistant.

3. Working in symbiosis with AI

The key is finding your own way of engaging with AI and stopping to overthink it. Symbiosis with AI is a combination of the scenarios in which you use AI and the ways you feed it digital inputs: dictation, external data, images, different formats.

Three quick tips:

Let AI create briefs, whether as a prompt for another AI task or as a project outline for a colleague
Generate multiple output formats at once: a post, an email, a summary
Give AI a goal instead of a specific task: “What would you do in my position to achieve this?”

Advanced tip: I have a prompt called “Conductor of Expertise.” For any task, AI first identifies five different experts in the relevant field, and then has them talk, discuss, and argue with each other. That’s how I arrived at the name “Superpowered Professional” for one of my core concepts.

4. Building your second brain

Easily one of the areas with the greatest potential.

Through intensive work with AI I’ve found you only need two things: a prompt and context. The prompt is the instruction, the context is your data. If you want AI to write emails the way you do, you need to instruct it and have your communication style documented. If you want automation, you need documented workflows and checklists. The better you document your knowledge, the more prepared you’ll be for the era of AI agents.

Practical example: Using Cursor, I had AI organize all the materials from the Future AI Leader program: presentations, transcripts, notes, tips. In 10–15 minutes everything was neatly structured, and today finding anything takes a matter of seconds.

Tip for companies: Add a button to your website that lets anyone copy the page content directly into ChatGPT or another AI tool with a single click. It helps customers too: when they need to convince a manager to approve a purchase, they can throw it into AI and get better arguments than they’d come up with on their own.

5. Creating content with AI

The key is knowing how to define your own communication style.

Every participant of the Future AI Leader program leaves with a set of assistants configured to work just like them, writing posts, emails, messages. All of it done in a way that no one can tell AI was involved.

My approach: I dictate my thoughts or write in bullet points. AI handles 90% of the work and I refine the final 10%.

More advanced options: We’re experimenting with video creation, cutting short clips from long formats, and using avatars or voice for book promotion. A word of warning: many trainers show how easy it is to do this or that with an avatar. It’s not that simple. Always check whether the person actually uses these tools themselves. Only learn from people who practice what they teach.

6. Accelerating projects

This is my favorite topic. Virtually everything we do today is a project, and AI can be used in almost every part of it:

Deep Research – AI scans dozens or hundreds of sources and produces a structured summary
Combining materials – original document + customer feedback + a few dictated notes = a new project or brief for colleagues
Information analysis – without needing a data analyst
Prototypes – of apps and services; just have a conversation with AI, drop in a transcript, create a brief, and have it code a prototype

7. Building digital projects and vibe coding

“Vibe coding” now exists: you describe what you want, AI writes the code. Kids are doing it. CEOs of billion-dollar companies are doing it. A 90-year-old pastor built an app for his congregation.

Reality check: it works well for simpler tools, utilities, and prototyping. Concrete examples from practice: cleaning data from five different spreadsheets, extracting 30 minutes of my voice from a podcast, personalizing contact lists.

Anything I’d assign to a colleague, I now assign to AI in plain language. The AI mindset is 90% mindset, 10% skill.

8. Working with data

It’s far simpler than it used to be.

Example: I wanted to understand how much I was using Google over time. I exported three months of browser history. Two prompts to AI produced a clean report. It turned out that in early February I was using Google 15 times more often than AI tools. Now I’m roughly 1:1.

The process: upload 3,700 rows to AI, ask questions, analyze, follow up. Once you know the basics, specifically what format to use and how to ask, it works reliably.

More advanced use: Cursor built me a script that merged contacts from 9 different sources, 30,000 contacts in total. I described what I wanted, AI wrote the script, and it was done.

9. Automation

Simple examples: analyzing a spreadsheet of opportunities in 10 minutes, updating contact records based on publicly available information.

More advanced: I had nearly 18,000 LinkedIn contacts and wanted to add department category, language, gender, and whether each person manages a team. I discussed the approach with AI, it created a brief, I handed it to a second AI. It worked for 8 hours with 99.9% accuracy. The entire operation cost $1.90.

10. AI trends and your role

This is the most important topic of all. It’s not about tools. It’s about how you think about your role. What will I do in the AI era? How do I build competence? How do I use AI to earn more, as an entrepreneur or as an employee?

It’s about using AI for better work, scaling, earning more with less, and above all, doing more of what we actually enjoy. One chapter of my book is called “The Golden Age of Creators.” That’s what all of this is really about.

Where to Start?

Mastering artificial intelligence is a matter of trial, experimentation, and learning. Many people know they’ll need to change how they work or learn something new, but most keep putting it off.

You won’t develop an AI mindset by watching webinars or asking colleagues. You develop it by working on real things. Pin an AI tool to your browser and keep trying. What you do between learning sessions matters more than the sessions themselves. Without that, nothing will stick.

If you want to reach a truly advanced level, take a look at the Future AI Leader program and stay tuned for when the next cohort opens.

The sooner you start, the bigger the head start. The future is already here. Don’t wait.