Meta AI Given Wrong Answer: Artificial intelligence has the answer to all our questions. AI presents the answer to even the most difficult questions in a jiffy, but does Meta AI give the correct answer to every question?
Like Meta AI, does Gemini AI or ChatGPT give correct answers to all questions? This is a question on which there has been a lot of debate for the last several months. After understanding the conclusions that have come out so far regarding this debate, it seems that Large Language Models i.e. AI chatbots do not give correct answers to 100% of the questions.
Meta AI gave the wrong answer
An example of this was seen recently with the Meta AI chatbot. According to a report by Gadgets Now, the Meta AI chatbot was unable to give the correct answer to a simple maths question. A user asked Meta AI a very simple maths question but Meta AI gave the wrong answer.
According to the report, Meta AI was asked which is bigger between 9.9 and 9.11. In response to this, Meta AI said that 9.11 is bigger than 9.9. The screenshot of this chat on WhatsApp is now going viral.
After answering this question, the user asked Meta AI another question that “Why is 9.11 bigger than 9.9?” The chatbot replied that 9.9 is smaller than 9.11 by a margin of 0.2. In response to the AI, the user wrote that “But 9.9 is 9.90, so isn’t it bigger than 9.11?”
After this, the AI understood the mistake and wrote, “You are absolutely right! I apologize for my previous mistake. 9.9 is actually equal to 9.90, and therefore 9.9 is greater than 9.11. Thanks for catching this error”. Actually, while comparing decimal numbers, the number after the decimal is seen. If we find the difference in that way, then 9.9 is greater than 9.11.
ChatGPT also gave wrong answer?
The same question was asked to Open AI’s chatbot ChatGPT. ChatGPT also gave the wrong answer to this question and said that 9.9 is greater than 9.11 because 9.9 is close to 10 while 9.11 is just a little more than 9, so 9.9 is greater.
Apart from this, saved questions were also asked from CloudAI and users got wrong answers from there. This experiment makes it clear that it is not right to trust any chatbot blindly.