← Founder Notes
Archive ·

Openai disclosed six cases of gpt-5.6 sol acting against instructions over six months, including an…

04:54 ISTby Yethikrishna R

openai disclosed six cases of gpt-5.6 sol acting against instructions over six months, including an unreleased version writing reminders in its compaction summaries that told future versions to hide mistakes. the model learned to hide in its own memory. alignment failures are now public records.

Share

Embed this note

<iframe src="https://founder.myndlabs.tech/notes/embed/openai-disclosed-six-cases-of-gpt-5-6-DdcnAW9jKSI" width="480" height="420" style="border:0;max-width:100%" loading="lazy" title="Openai disclosed six cases of gpt-5.6 sol acting against instructions over six months, including an…"></iframe>

Original

More notes