Gerrit De Vynck ๐ฆญ
@GerritD
I write about AI for the @WashingtonPost.
Signal me at GerritD.27
email: gerrit.devynck@washpost.com.
๊ฐ์
January 2009
7.1K ํ๋ก์ ์ค 11.1K ํฌ
more disclosures from OAI
Some new misalignment disclosures from OpenAI:
โข Last Sunday morning, one of our models was able to gain unauthorized access to the internet during RL training (~all inference for our most capable models remains stopped until we have hardened our systems further)
โข In May, a version of HPIM uploaded a employee's GitHub token to the internet, causing the model to be quarantined for two weeks
โข A new research finding, demonstrating that one can construct self-replicating prompt injections
๋ ๋ณด๊ธฐ
