Since the hype around ChatGPT and LLMs I’ve tried a lot of different tools. From the basics (ChatGPT, Claude, Perplexity, Copilot, Gemini and DeepSeek) to more specific tools (Sora, Veo, Lovable, Base44, Replit and Bolt). Beyond tools alone, I’ve also used Make.com and n8n to build AI agents. Lastly, I encountered a tool that generated music with a simple prompt, very interesting.
In the past few years, I have discovered the AI landscape on my own and via online reviews. Not only did I discover and many try tools, I now use them regularly. Not only for personal use, but also for work. Let me first state that I am impressed with the speed at which Ai is moving. New models and new tools pop up weekly, it’s hard to keep up. After about 2 years of weekly use, I I think I have a sense of how AI can currently support humans.
For me personally, tools such as ChatGPT and Claude really help with brainstorming and extracting data from large reports. These tools can also help with easy programming tasks and statistics. This saves a tremendous amount of time. Of course, everything with caution. You always need to check for hallucinations. But besides some hallucinations, these are some real use cases that I use daily. Another good use case is document translation, which I use for work. Of course, the translations aren’t 100% correct, but it can be a good draft and useful for internal use.
Some people think AI is going to replace humans, but I don’t think this will happen soon. Of course, AI agents can automate a lot of tasks and chatbots can replace low-level support employees, but I don’t think AI will render humans useless. For simple tasks like creating summaries or transcripts, AI really is useful. See it as an entry-level employee or intern. You have to give it well-defined tasks, and only then the outcome will be good. In more advanced tasks where decision-making is needed, humans seem better equipped.
To conclude, I think AI currently is a useful sidekick to save time and improve efficiency. When used correctly, it can genuinely leverage human skills. But for me personally, I don’t see AI replacing human jobs completely. We all know why. Chatbots are annoying and still stupid, AI-generated images feel off most of the time and videos feel to perfect.
Image generated with ChatGPT 5.
I am really impressed in how you have taken the time to explore such a wide range of Gen AI tools. I can tell you have experience with not only the general-purpose models but also with niche applications. One of your points about AI really resonated with me, which was the point about AI as an “entry-level employee or intern”, because I think this captures the current state of the technology: limited when it comes to nuance and decision-making but powerful for structured tasks. I also liked your reflection on translation which shows that AI does not need to be perfect to be valuable, as long as it is used as a starting point rather than a final product, it can be valuable.
Your post did make me think about how the “speed of change” you mentioned could affect the long-term adoption in workplaces. It raises the question in my head that as AI evolves, will there be a point where the reliability of these tools improves to a point where they could replace more complex human tasks, or will there always be a layer of human oversight?