AI is meant to make people 200% more productive
In this way, they can save more time to finish those tasks that machines can’t replace human work, or they can work on improving the AI models to make them more capable. The wide concern of how many people may lose their jobs due to the development of AI reflects people’s understanding and expectations of AI. There’s no doubt that the thriving of AI has greatly influenced everyone’s life in a better way.
A very intuitive example of how AI performs well in repetitive work is chatbots. Guess we all have the experience of turning to customer service for help. Whenever it’s online chatting or calling from the phone, you will be greeted by AI chatbots first. They will try their best to understand your questions and give corresponding solutions. If they can’t solve your problem properly, you will be transferred to a real person. What chatbots are capable of is based on the collection of questions and answers collected from past conversations. Thus they should be able to deal with common concerns. Chatbots can offer customers fast, accurate, and timely support. Not only do they save human energy answering these “dumb” questions, but they also improve customers' overall experience. They also make collecting relevant data from customers a lot easier. In return, these data can be used to train AI chatbots to make them “smarter”.
Personally, I think AI like ChatGPT can serve as great search engines. Powered by natural language processing (NLP), AI can understand your questions better. No matter if you are only providing keywords or the whole sentence, AI can effectively extract the points in your question. While for search engines, sometimes it takes some time to find the best combination of keywords that leads to the best solution. AI also provides more accurate answers with less time you need to put into searching. For questions so certain like what does one plus one equals to, you can trust AI that it will give a correct answer. For other scenarios, let’s say I met an uncommon bug that only a few communities are discussing about it. In their discussions, there are nonsense, large portion of codes, failed tries, and of course, the answer. I have to go through tens of pages, try each solution to see if it actually works. But AI can provide me with the solution directly. If not this powerful, it can still provide me with a list of items that I should look into that might be related to the bug. It saves me so much time and energy.
Talking about how AI improves productivity, I can’t miss how AI can help boost creative work. Just as I mentioned in my previous post, in the area of image and design generation, AI is good at jumping out of the box, seamlessly blending styles from different artworks, or generating imaginative scenes that go against conventional imagination. AI can also generate new content, like poems, articles, or even music. Even though they are not able to give complete and real human-like pieces, they can serve as nice starting points or inspirations to artists, allowing them to put more focus on refining and adding personal flavors.
Of course, there are some side effects coming with AI. First, the initial implementation will require significant investment. It’s difficult to start from scratch if you want to train your own AI model. The results AI gives may not be that satisfiable. Lack of understanding of input contexts, inherent bias in training data, or creativity limited by input dataset, can all lead to users wasting time telling how reliable the answer is and generating answers over and over again. It will just be a waste of time if the user is clear about how well the result is. If not, for example, the user is looking for the answer to a complicated math problem, then faulty result can cause a lot of trouble.
In conclusion, though AI holds great potential of revolutionizing the way people live, it’s important that we realize the limitations and keep on working on the weak spots.