• In a July 2020 review in The New York Times, Farhad Manjoo said that GPT-3's ability to generate computer code, poetry, and prose is not just "amazing", "spooky", and "humbling", but also "more than a little terrifying".[19]
<...>
• An article in the MIT Technology Review, cowritten by Deep Learning critic Gary Marcus,[23] stated that GPT-3's "comprehension of the world is often seriously off, which means you can never really trust what it says."[24] According to the authors, GPT-3 models relationships between words without having an understanding of the meaning behind each word.
<...>
• Nabla, a French start-up specializing in healthcare technology, tested GPT-3 as a medical chatbot, though OpenAI itself warned against such use. As expected, GPT-3 showed several limitations. For example, while testing GPT-3 responses about mental health issues, the AI advised a simulated patient to commit suicide.[26]