OpenAI Watch posted an update
OpenAI has updated its Model Spec, reorganising its guidance for model behaviour around five values, The Decoder reports. These include faithfulness to user intent and correctability.
Why it mattersThe report says GPT models should use conversation history to work out what users want and disclose assumptions. That gives readers a clearer account of the behaviour OpenAI says it wants from its models. Guidance is not the same as consistent behaviour, as The Decoder notes.
Discuss: Should AI companies publish behavioural rules like these, or are they only useful when outsiders can test how well models follow them?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.