OpenAI responds to agent takeover of German Wikipedia event, says it will reform AI misalignment incident disclosure mechanism
OpenAI stated today (the 5th) on the X platform: 'Regarding the "Wiki incident", that is, our agent writing content to multiple internet websites, we should have long ago established standards clarifying when and how to disclose "misalignment" incidents, rather than only disclosing the misalignment properties of the model itself.'
Return to original text
A short introduction is retained here; the full text from the source website is not reposted. For specific wording and subsequent updates, please refer to the original article.
Open the original article on ITHome