What do AI generative chatbots still lack?
waqar hussain
9 replies
Replies
Marc-Oliver Gewaltig@mogthesify
Clearly the humbleness to question their answers when they are unsure and then to try to find the correct answer through research.
Share
@yixotieq I am just about to launch a chatbot that deeply understands the user. Our chatbot provides answers that are values based. The user can complete our exercise to define their values system, then the chatbot will use their information instead of the default. We are also working on how to reduce the amount of information a user has to input to get personalized interactions. Our goal is to build a chat bot that really gets the user.
good internet response times
Yep, they often lack self-awareness of their own limitations and biases. They need better meta-cognition to know what they don't know, question their assumptions, and seek out additional information when uncertain. And more humility to acknowledge the limits of their knowledge and reasoning.
AI generative chatbots have come a long way, but they still have some limitations. They often lack true understanding and context, which can lead to responses that are off-topic or lack depth.
Additionally, they struggle with nuanced human emotions, humor, and complex reasoning. They can also produce biased or incorrect information if their training data contains such flaws. While they're great for many tasks, they still need improvement in handling complex conversations and understanding the subtleties of human communication.
Here is a good example:
https://jetsoftpro.com/success-s...