How 2 Science

Artificial intelligence (AI) rapidly transforms how we work, think, and create. Tasks that once required years of training can now be assisted—or even automated—by just a few lines of prompts. This shift is unlike previous technological changes; it doesn’t just enhance our tools—it challenges the very nature of expertise. So, how are scientists being affected by this disruption? And more importantly, how do we adapt to this new world?

The Rise of Accessible AI

AI has been used in fields like engineering, computer science, and chemistry for decades—and more recently, it has expanded into biology and the social sciences. So why has it only captured the public’s attention in the past few years?

Much of this shift can be credited to ChatGPT. While large language models (LLMs) have existed for years, their sudden accessibility—requiring nothing more than an internet connection—pushed AI into the public spotlight. As these tools reached a broader audience, a competitive race emerged between companies, each striving to outdo the others in speed, accuracy, capabilities, and user experience.

For users, this explosion of innovation has opened doors to a wide range of capabilities—from generating written content and images to writing code and building applications. AI seems to be evolving faster than we can keep up. So where does this leave us as scientists? How can we use these tools meaningfully in our daily work and research?

AI Tools Hiding in Plain Sight

AI tools are already embedded in many of the apps you use every day—often without you even noticing. Take Gmail, for example. If you use it, you’ve likely encountered Gemini, Google’s integrated AI assistant. It can summarize emails, suggest replies, and do much more—features many users barely scratch the surface of. And if something as routine as email is now powered by AI, it’s easy to imagine how deeply these tools are influencing the software we rely on for scientific research.

So what are these tools, really? How do we use them effectively—and is there a cost? As with most powerful technologies, the answer isn’t straightforward.

One AI Tool to Rule Your Workflow

The AI landscape is overflowing with specialized tools. Need to generate a video? Kling has you covered. Want a lifelike narrator? ElevenLabs does that. Every week, a new startup releases another narrowly focused AI product—almost always behind a paywall. If you’re looking to use AI for a range of tasks, the cost of these subscriptions can quickly spiral out of control.

But you may not need all of them. A few powerful, general-purpose tools—like ChatGPT—offer a surprisingly wide array of features. With a single subscription, you can draft documents, generate images, summarize articles, write code, and even create short videos. While some features are still maturing, their flexibility and breadth make them extremely valuable.

That’s why I recommend focusing on just two or three versatile AI tools, rather than spreading your budget across dozens of niche platforms. In the next section, I’ll share the tools I use today—though like the technology itself, my toolkit continues to evolve.

My Personal AI Toolkit (and Why I Use It)

The first tool I rely on is ChatGPT. While I briefly experimented with alternatives like Gemini, I found myself returning to ChatGPT for its balance of accuracy, flexibility, and overall quality. It consistently delivers across a wide range of tasks, from writing and coding to brainstorming and editing.

The second tool I use is Grammarly. Since I regularly write logs, documents, and—soon—research papers, having clear, professional language is essential. Grammarly integrates seamlessly with both Chrome and Word, catching errors as I write and helping me refine my tone. I can even tailor the writing style: formal for academic reports, or more conversational for blog posts like this one. It doesn’t just correct my mistakes—it helps me improve my English as I go.

The third tool is Speechify. Since my commute to the university can take over an hour, I started using that time more productively by listening to research papers and educational articles. Speechify converts almost any text to audio and integrates easily with Chrome. With just a click, I can listen to papers on the go—making the most of my time and continuing to learn, even outside the lab.

Final Thoughts

There’s far more to say about AI than what I’ve covered here—from integrating it into your workflow to comparing the ever-growing list of tools. But this post wasn’t meant to be comprehensive. My goal was simply to share my personal experience and encourage others—especially fellow scientists—to explore these tools for themselves.

The best way to find out what works for you is simple: try it. There’s no substitute for hands-on experience. Some researchers may prefer Gemini over ChatGPT. Others might rely on a combination of ten specialized tools tailored to their field.

Ultimately, it’s about finding what fits your workflow—and adapting both the tools and your habits to make the most of them.

I hope this post has encouraged you to see AI in a new light—and inspired you, even just a little, to explore its potential more fully in your scientific work.