I first came across ChatGPT at the beginning of 2023, during my bachelor’s studies. At that time, I had to write an essay and felt stuck on how to structure it properly. Out of curiosity, I decided to test ChatGPT, which was still relatively new to most students. To my surprise, it gave me a clear outline and suggestions for different angles I could take. It didn’t “do the work” for me, but it provided a helpful framework that allowed me to organize my thoughts more efficiently. That first encounter made me realize that AI could be more than a fun experiment, it could actually serve as a learning partner.
Since then, my use of AI tools has grown in both frequency and diversity. ChatGPT became a regular study companion for brainstorming, drafting and even getting feedback on assignments. When I was working on more demanding projects, such as my thesis, it also helped me with learning Python. For someone without a coding background, being able to ask ChatGPT to explain scripts in simple terms or generate a basic function was very useful. It allowed me to focus on the analysis and interpretation rather than getting stuck on the technical details.
Of course, I’ve also learned the limitations the hard way. ChatGPT can sometimes “hallucinate,” meaning that it produces information that sounds very confident but is actually completely inaccurate (Hua et al., 2023). This forced me to develop a cautious approach, treating it as a supportive tool but not as the final authority. It has been a reminder that, while AI can accelerate the learning and problem-solving, critical thinking and fact-checking still are very essential.
Beyond ChatGPT, I’ve explored other forms of generative AI as well. One notable example was using Lookx.ai for work, a platform that generates 3D designs (LookX AI Cloud, 2025). We used it to model tiny houses, which was both exciting and practical. Instead of relying on time-consuming manual design, the tool allowed us to quickly test out visual concepts and bring abstract ideas to life. Seeing how AI could streamline creative design processes made me realize just how much these technologies cqn do, which extends beyond education into professional and even artistic domains.
Looking back, my experience with generative AI has been a mix of excitement, productivity and careful reflection. I see it as an “augmenter” of human capability rather than a replacement. It has helped me overcome creative blocks, learn faster and to test ideas that would have otherwise taken much longer. At the same time, it has also taught me to remain critical, aware of biases and inaccuracies and to be responsible in how I apply these tools.
For the future, I think personalization will be key. Current tools are very powerful but still generic, they don’t always capture the individual preferences or working styles. A truly transformative step would be systems that adapt to each user over time, much like a colleague who understands your way of thinking. That’s where I see the real potential of AI, not just as a smart assistant, but as a trusted collaborator.
References:
Hua, S., Jin, S., & Jiang, S. (2023). The Limitations and Ethical Considerations of ChatGPT. Data Intelligence, 6(1), 201–239. https://doi.org/10.1162/dint_a_00243
LookX AI Cloud. (2025). https://www.lookx.ai/
Image credit: https://silva.builders/
From Essays to 3D Houses: My Journey with Generative AI
3
October
2025
Hi Hollie 🙂
What led me to read your post is the 3D Design example you brought, which was a real standout for me!
I’d never thought about how a fast, initial render can be genuinely useful to materialise an idea, in some cases even more quickly and accurately than a hand-drawn sketch.
Your reflection also nailed a central dilemma I’ve been grappling with. When AI helps us overcome a creative block, is it actually unlocking something within us, or is it just performing the creative task we couldn’t? Your experience suggests it’s a partner, but I think it’s a crucial question for all of us to keep asking: are we using this tool to enhance our creativity, or are we letting it become a creative prosthesis?
I agree that the future lies in personalised systems that act as true collaborators. Perhaps then, the line between tool and partner will become clearer. Thanks for sparking these reflections!