Googles large language model Gemini has come up with problematic responses related to PM Modi. ChatGPT was generating gibberish after experiencing a crash due to a DDoS attack. The incidents have led to accusations of bias and potential violation of the IT rules. ET Prime explains the algorithms behind the LLMs and decodes their reliability and ethical implications.