Coding advice website Stack Overflow has temporarily banned responses generated by OpenAI's ChatGPT AI program, citing "a high rate" of incorrect answers. ChatGPT, based on the GPT-3.5 language model, can answer questions about coding problems but often produces "plausible-sounding but incorrect or nonsensical answers," according to OpenAI.
More:
- ChatGPT can write new code, fix errors, and ask for additional content in coding problems.
- On Stack Overflow, users turned to ChatGPT to supply many answers without first checking if they were factually correct, according to ZDNet.
- In a statement, Stack Overflow said it has temporarily banned ChatGPT as the AI's average rate of correct answers remains "too low."
- The organization is concerned that the responses could be "substantially harmful." It plans to hold a staff discussion and issue a permanent decision on ChatGPT use at a later date.
See the full story here: https://www.vice.com/en/article/wxnaem/stack-overflow-bans-chatgpt-for-constantly-giving-wrong-answers
The text above is from the Inside AI newsletter