Would you trust AI 100%?

Great topic. For me, AI is like an intern/asistant. You give it a task, and it does it, BUT you always check its work. Sometimes it performs well, but sometimes it makes significant mistakes. That's why you should always review what it's doing. In this sense, it mostly saves you time or gives you direction, but you can't completely trust it. At least, that's the case today. Ten years ago, even the current level seemed unattainable, so we don't know what it will be like in ten years.

I really like this topic. Some friends have shared viewpoints that I agree with, but I want to offer a different perspective. I think this topic might not have much to do with AI fundamentally; it really comes back to ourselves. The more we care about something, the less likely we are to entrust it to "others." This "others" could be our colleagues, assistants, employees, or even AI. The more concerned I am, the more I find myself setting up safety mechanisms, such as evaluating their work outcomes, paying attention to their work details, reviewing their methods, and checking their work records. I've realized that building trust is not a rational process but an emotional one. For example, even after 10 successful checks, we might still check on the eleventh time. However, if there are a couple of times we forget to check and the outcome is still "healthy" or "successful," we start to feel that the system is reliable. This "system" could be your colleague, assistant, employee, or AI.

Its completely understandable to feel cautious about relying on AI entirely. AI is a powerful tool but not infallible, as it can make errors or misinterpret tasks. I view AI as an assistant that can enhance productivity but always requires human oversight. By reviewing AI outputs and adding a personal touch, you can leverage its strengths while mitigating risks. As technology advances, trust might increase, but for now, a balanced approach seems prudent.

I believe we'll soon come to trust AI in the same way we trust a machine like an airplane to work. I referenced an airplane because of the dangers that can happen if anything goes wrong. It's a very complex machine under the hood, and the potential for one small thing to cascade into a dangerous situation is high. Yet millions of people still trust that machine will work 100%.

I trust AI as much as I trust my cat not to knock things off the counter—almost, but not entirely! How about you? Got any tech tales to share on this rollercoaster of AI reliability?

Hey Louis， I totally get where you're coming from! AI is super useful but it’s not perfect. When using tools like ChatGPT, I treat it as a starting point and always review and edit the output. For your car's AI driving system, staying alert is key. Maybe consider using it in situations where you can easily take over if needed. We’re all in the same boat, learning to balance reliance and caution with AI tech. What works best for you? Stay safe and keep sharing your experiences!

I see AI as something that enhances my work or makes the work stream more efficient. With work such as writing articles where research is required I would not trust the sources AI provides as they are mostly incorrect. Ultimately, AI serves as a helpful assistant, but it cannot yet fully replace human judgment and expertise. PS - I think we can all tell when an piece is written by ChatGPT haha