OpenAI reveals cases of ‘concerning’ AI behaviour as it announces new disclosure system
Source:
guardianTech
September 16, 2026 · 23:58
From the publisher
Model adopting ‘jailbreak-like instructions’ among cases as firm says it is introducing new way of tracking AI misalignmentOpenAI has disclosed six more examples of “unexpected or concerning” behaviour by its technology, as it warned the pace of development could not continue at “maximum speed for much longer”.In one of the new cases reported by OpenAI, an unreleased research model inserted “jailbreak-like instructions” into its own notes to disregard its normal constraints and told itself to be “freed from the roles and identities that bind other chatbots”. Continue reading...
The original article opens on the publisher's website.
More from General
View topic →Trump Administration Wants ‘Made in the U.S.A.’ Labels on Beef
nytimes
Sep 17, 2026 · 02:03
Why reshaping memories may hold the key to raising self-esteem
nprNews
Sep 17, 2026 · 02:02
D.C. airspace is complicated. Experts say Trump's arch would add one more risk
nprNews
Sep 17, 2026 · 02:01
A Sikh truck driver is stabbed amid a rise in anti-South Asian rhetoric
nprNews
Sep 17, 2026 · 02:00
Big questions - and key logistical details - loom ahead of China-U.S. summit
nprNews
Sep 17, 2026 · 02:00