Don't put blind trust in ChatGPT, warns Sam Altman
What's the story
Sam Altman, the CEO of OpenAI, has warned users against putting blind trust in ChatGPT. Speaking on the first episode of the official OpenAI podcast, he said, "People have a very high degree of trust in ChatGPT, which is interesting because AI hallucinates." "It should be the tech that you don't trust that much," he added. The term 'hallucination' refers to when an Artificial Intelligence (AI) model generates information that is inaccurate or misleading but presents it with confidence.
AI reliability
Hallucination can be caused by several factors
Altman further stressed on the podcast that "it's not super reliable... we need to be honest about that." He explained that hallucination in AI can be caused by several factors, including biased training data, lack of grounding in real-world knowledge, pressure to always respond, and predictive text generation. Despite its widespread use, no major AI company has claimed its chatbots are free from this issue.
Future outlook
My children will never be smarter than AI: Altman
Altman also reiterated his previous prediction that his children will never be smarter than AI. However, he added, "But they will grow up like vastly more capable than we grew up and able to do things that would just, we cannot imagine." This statement highlights the potential future capabilities of humans in an increasingly AI-driven world.
Monetization strategy
Ads on ChatGPT?
When asked about the possibility of ads coming to ChatGPT, Altman said he wasn't totally against it. He said, "I'm not totally against it. I can point to areas where I like ads. I think ads on Instagram, kinda cool. I bought a bunch of stuff from them." However, he also noted that implementing such a model would require careful consideration not to disrupt user experience or compromise LLM's output quality.