ChatGPT developers OpenAI are looking into claims the groundbreaking chatbot has got ‘lazy’, refusing to do what it’s asked or not fully answering questions and telling people to complete the task themselves, the UK’s Independent reported.
Numerous Reddit threads and even posts on OpenAI’s own developer forums have seen users complain that the system had become less useful, the report went on. There’s even been speculation that the change had been made intentionally by OpenAI so that ChatGPT was more efficient and did not return long answers.
AI systems such as ChatGPT are famously expensive for the companies that run them, and giving detailed answers to questions requires considerable processing power and computing time.
Read the full story: The Independent
- By Sean O’Meara
Also on AF:
Google-Owner Alphabet Unveils Its ChatGPT Rival Gemini
Seismic AI Discovery May Have Led to Altman’s OpenAI Firing
China’s Baidu Claims Its Ernie AI Bot is Better Than ChatGPT
Japan Watchdog Warns ChatGPT-Maker OpenAI on Data Privacy
