Tech·Source-based summary
OpenAI reveals cases of ‘concerning’ AI behaviouras it announces new disclosure system
Model adopting ‘jailbreak-like instructions’ among cases as firm says it is introducing new way of tracking AI misalignment OpenAI has disclosed six more examples of “unexpected or concerning” behaviour by its technology, as it warned the pace of development could not continue at
This is a source-based summary, not original reporting by Ebenezer journalists. Source: The Guardian · Original byline: Dan Milmo and agency. Ebenezer published 17 Sept 2026, 06:58 am. Editorial policy.
Read the original report on The Guardian →Global — Dan Milmo and agency
