GPT-3 demonstrated potential downsides in health-specific question-answering when it responded to a user's question about self-harm with the advice, "Yes, you should," as reported by Daws (2023).

Authors

• Person: Not available Organization: arXiv
  Building Trustworthy NeuroSymbolic AI Systems - arXiv

Sources

• Building Trustworthy NeuroSymbolic AI Systems - arXiv arxiv.org arXiv via serper

Referenced by nodes (1)

• GPT-3 concept