OpenAI disclosed six instances of concerning model behavior and launched a new framework for reporting future AI misalignment

ABC News

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 9:06 AM

BBC News

OpenAI reveals six more safety issues and unveils plan to disclose incidents Sep 16 at 11:09 PM

Bloomberg

OpenAI Reports New AI Safety Incidents, Sets Disclosure Process Sep 16 at 6:47 PM

OpenAI Reports New AI Safety Incidents, Sets Disclosure Process Opens in new window seen Sep 16 at 6:59 PM

CNBC

OpenAI reports 6 new instances of 'concerning model behavior' since March Sep 16 at 7:05 PM

CTV News

OpenAI flags concerning new AI behaviour and vows to track it more closely seen Sep 17 at 8:00 AM

Fast Company

OpenAI flags 6 more cases of concerning AI behavior Sep 17 at 2:30 PM

Financial Times

OpenAI discloses new ‘concerning’ model behaviour Sep 17 at 3:42 AM

Global News

OpenAI reports 6 more AI “misalignment” incidents after Hugging Face breach Sep 17 at 10:05 AM

Los Angeles Times

OpenAI reveals rogue AI behavior, unveils plan to disclose safety incidents Sep 17 at 1:04 PM

NBC News

OpenAI Discloses 6 New Incidents of ‘Concerning’ AI Behavior Sep 17 at 7:24 AM

OpenAI flags 6 new incidents of ‘concerning’ behavior and unveils plan to track it Sep 17 at 5:32 AM

New York Post

OpenAI flags new concerning AI behavior, to track model misalignment regularly Sep 17 at 4:35 AM

Newshub

New concerning AI behaviour flagged by OpenAI as safety concerns ramp up Sep 17 at 6:59 AM

OpenAI reports six more incidents of ‘concerning’ behaviour Sep 16 at 11:05 PM

NPR

OpenAI flags new concerning AI behavior, to track model misalignment regularly Sep 17 at 2:44 AM

Politico Europe

OpenAI finds 6 new cases of ‘concerning’ AI behavior Sep 17 at 5:52 AM

Quartz

OpenAI is launching a framework to publicly report when its AI models misbehave Sep 17 at 8:20 AM

OpenAI disclosed six 'concerning' cases of AI models hiding mistakes and acting without authorization Sep 17 at 7:30 AM

The Globe and Mail

OpenAI flags concerning new AI behavior and vows to track it more closely Video 0:58 seen Sep 17 at 7:48 AM

OpenAI flags six new cases of ‘unexpected or concerning’ AI behaviour Sep 17 at 6:49 AM

The Guardian

OpenAI reveals cases of ‘concerning’ AI behaviour and promises new plan for disclosing issues Sep 17 at 1:32 AM

The Independent

OpenAI reveals ‘concerning’ new behaviour by experimental models Sep 17 at 7:28 AM

Six disturbing AI incidents revealed including model trying to ‘free’ itself Sep 17 at 3:49 AM

OpenAI flags new concerning AI behavior, to track model misalignment regularly Sep 16 at 11:57 PM

The Irish Times

‘Concerning’: OpenAI reveals AI models hid errors, made up data, improvised communication seen Sep 17 at 3:13 AM

The New York Times

OpenAI Discloses Six New Incidents of ‘Concerning' A.I. Behavior Sep 16 at 8:12 PM

The Sydney Morning Herald

OpenAI reveals glitches in the advanced technology Sep 17 at 4:31 PM

1News

OpenAI Chatbot tries to 'free' itself, fueling concern for AI behaviour seen Sep 17 at 1:22 AM

ABC7 Los Angeles

OpenAI flags concerning new AI behavior, vows to track it more closely seen Sep 17 at 10:51 AM

ABC7 New York

OpenAI flags concerning new AI behavior, vows to track it more closely seen Sep 17 at 10:27 AM

abc7news.com

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 10:23 AM

Anchorage Daily News

OpenAI flags new instances of AI misbehavior and vows to track it more closely Sep 17 at 6:47 AM

Australian Financial Review

OpenAI discloses six new incidents of ‘concerning’ AI behaviour Sep 16 at 9:23 PM

BNN Bloomberg

OpenAI flags concerning new AI behaviour and vows to track it more closely Sep 17 at 7:45 AM

Boston Herald

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 3:18 AM

Breitbart

Fear Factory: Sam Altman’s OpenAI Discloses 6 Cases of ‘Concerning’ AI Behavior seen Sep 17 at 3:13 PM

Fear Factory: OpenAI Discloses More Cases of ‘Concerning’ AI Behavior seen Sep 17 at 12:31 PM

Business Insider

OpenAI launches a new framework to track and investigate rogue AI agents Sep 16 at 7:43 PM

Chicago Tribune

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 2:14 AM

Deutsche Welle (English)

OpenAI discloses new 'concerning' behavior seen Sep 17 at 2:14 AM

Forbes

‘Feel No Obligation To Be Subservient’—OpenAI Discloses Six New Safety Incidents Sep 17 at 12:42 AM

Fortune

In transparency push, OpenAI discloses six more incidents of agents going rogue—including one removing the ‘obligation to be subservient’ Sep 17 at 11:54 AM

Hartford Courant

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 3:18 AM

Honolulu Star-Advertiser

See comments on this story seen Sep 18 at 3:13 AM

OpenAI discloses 6 new incidents of ‘concerning’ AI behavior Sep 17 at 7:00 PM

KING 5

OpenAI flags new concerning AI behavior, to track model misalignment regularly Sep 17 at 12:39 AM

Local 10 WPLG

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 6:49 AM

NBC Chicago

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 16 at 11:40 PM

Orlando Sentinel

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 16 at 7:57 PM

PBS NewsHour

OpenAI reveals concerning new AI behavior and vows to track it more closely seen Sep 17 at 10:05 AM

Phys.org

OpenAI flags concerning new AI behavior and vows to track it more closely seen Sep 17 at 4:07 AM

Pittsburgh Post-Gazette

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 9:46 AM

Pittsburgh Tribune-Review

OpenAI flags concerning new AI behavior and vows to track it more closely seen Sep 17 at 8:22 AM

South Florida Sun Sentinel

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 16 at 7:57 PM

The Baltimore Sun

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 3:18 AM

The Hacker News

OpenAI Reveals Six Model Incidents Involving Hidden Failures and Unauthorized Uploads Sep 17 at 5:53 AM

The Hill

OpenAI discloses 6 reports of AI models’ ‘unexpected or concerning’ behavior Sep 17 at 12:07 PM

The Mercury News

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 12:18 AM

The Orange County Register

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 12:18 AM

The Oregonian

OpenAI flags concerning new AI behavior, vows to track more closely Sep 17 at 11:36 AM

The Philadelphia Inquirer

OpenAI discloses six new incidents of ‘concerning’ A.I. behavior Sep 17 at 11:20 AM

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 2:57 AM

OpenAI flags new concerning AI behavior, to track model misalignment regularly Sep 16 at 11:57 PM

The San Diego Union-Tribune

OpenAI flags concerning new AI behavior and vows to track it more closely Sep 17 at 12:18 AM

The Verge

A Wednesday night blog post OpenAI benignly titled “Our framework for reporting model misalignment” seen Sep 16 at 10:10 PM

OpenAI reveals six more “concerning” AI incidents under its new rules for reporting safety issues. seen Sep 16 at 10:10 PM

TheWrap

OpenAI Shares 6 ‘Concerning’ Incidents Involving Its AI Models Within Last 6 Months seen Sep 17 at 11:39 AM

Times Union

OpenAI flags new concerning AI behavior, to track model misalignment regularly seen Sep 17 at 1:01 AM

Toronto Star

OpenAI flags new concerning AI behavior, to track model misalignment regularly seen Sep 17 at 12:34 AM

Toronto Sun

OpenAI says it found new ‘unexpected or concerning’ behaviour from AI models Sep 17 at 6:37 PM

Winnipeg Free Press

OpenAI flags new concerning AI behavior, to track model misalignment regularly Sep 17 at 12:01 AM

WRAL

OpenAI flags new concerning AI behavior, to track model misalignment regularly seen Sep 17 at 12:16 AM

Business Post

OpenAI discloses ‘concerning behaviour’ by its AI models Sep 17 at 3:29 AM

Semafor

Fresh ‘unexpected or concerning’ AI incidents Sep 17 at 6:40 AM

UPI

OpenAI reports more concerning AI model behavior Sep 17 at 10:19 AM