AI Can Still Give Wrong Information
Sam Altman, CEO of OpenAI and the creator of ChatGPT, has warned people not to blindly trust everything AI says. He said that even though people are using artificial intelligence more and more, AI still has the risk of giving wrong or made-up information. Many users, however, accept these answers without checking.
People Trust ChatGPT Too Much
In the first episode of OpenAI’s own podcast series, Sam Altman shared that ChatGPT has gained a lot of user trust. While that sounds good, he pointed out that sometimes ChatGPT gives answers that are not based on real facts. It can even explain things that don’t exist in reality — these are just made-up stories.
Sam Altman’s Own Experience with ChatGPT
Altman talked about his personal use of ChatGPT. He recently became a father, and during that time, he used ChatGPT often. From making a sleep routine for his baby to finding solutions for diaper rashes, he relied on the chatbot. However, he noticed that not all the answers were correct.
Read More: The Final Farewell of Manu Mia, the selfless gravedigger, who served love nearly 50 years
He said, “Even if AI is getting more useful, final decisions should always be made by humans.” He believes we are just at the beginning of using this powerful technology. If we are not careful now, trust in AI might grow faster than the actual ability of the technology.
Data Privacy and AI’s Future Plans
In the podcast, Altman also talked about other important issues like protecting user data, making AI systems more transparent, and the possible launch of ad-based revenue models in the future.
He said OpenAI is planning to add new features to ChatGPT, like “persistent memory” — which means the chatbot would be able to remember past interactions for a better user experience. But he promised that user trust and safety will always be the top priority.
Source: Prothom Alo
