Elon Musk is challenging OpenAI’s ChatGPT with his Grok chatbot, asserting that Grok 4 outperforms ChatGPT 5. Recently, a user shared an interaction with ChatGPT 5’s Thinking model, where it admitted, “I don’t know,” after a 34-second deliberation. Musk found this response “impressive.” This admission indicates a significant improvement in managing hallucinations—instances where AI generates inaccurate information confidently.
OpenAI has made strides to mitigate hallucinations since ChatGPT’s launch in November 2022, though a 10% chance still exists in ChatGPT 5. Despite advancements, OpenAI emphasizes that users should not rely solely on ChatGPT for information. ChatGPT head, Nick Turley, advises users to verify facts, highlighting that the model should serve as a secondary source rather than the primary authority. Since its August 7 release, ChatGPT 5 has gained many active users, complemented by the launch of ChatGPT Go, an affordable plan for Indian users.
Source link
