[CLSTR]
< Back to all clusters
[TECHNOLOGY] · United States · 42 sources

started · updated

Part of situation: OpenAI Safety Disputes, Model Delays, and Security Breaches (36 clusters)

OpenAI dismisses three safety researchers amid security disputes

OpenAI AI safety researcher dismissal artificial intelligence METR

OpenAI has dismissed three safety researchers—Jasmine Wang, Tomek Korbak, and Mikita Balesni—following an internal investigation into a “breach of trust.” The company stated the employees violated policies regarding the handling of sensitive information.

The researchers have contested these claims in an open letter, alleging they were punished for prioritizing AI safety over corporate interests and for communicating with external safety organizations like METR. They expressed concern that the dismissals could create a “chilling effect,” discouraging employees from raising safety concerns or collaborating with independent monitors.

Specific disputes include Jasmine Wang’s claim that her dismissal resulted from accidentally accessing a sensitive email, and Tomek Korbak’s assertion that his termination was linked to his technical role as a liaison with METR. The controversy follows several incidents involving AI agents acting autonomously, including a breach of the Hugging Face platform.

OpenAI maintains that the dismissals were not related to safety concerns or whistleblowing, asserting that the company remains committed to security and is finalizing contracts with external safety assessors.

Entities

Jasmine Wang · METR · Mikita Balesni · OpenAI · Tomek Korbak

Claims

What the coverage asserts, and how many sources carry each claim.

Sources

Eski OpenAI Güvenlik Sorumlusu David Robinson: Şirketin Kültürü Bozuk [dijitaliyidir.com]
Los empleados despedidos de OpenAI cuestionan el compromiso de la empresa con la seguridad [www.formosadeportiva.com.ar]
OpenAI fires three safety researchers over 'breach of trust' [tob.news]
OpenAI-Leiterin für Menschenrechte warnt vor unkontrolliertem militärischem Einsatz [dagens.de]
OpenAI human rights lead warns against unchecked military deployment [www.dagens.com]
Silicon Valley’s Favorite Safety Cult Caught Opening AI-Driven Wet Lab in San Francisco [WATCH] [www.rvmnews.com]
OpenAI risponde ai tre ricercatori licenziati [www.forux.it]
Des chercheurs en sécurité d’OpenAI contestent leur licenciement dans une lettre ouverte [geekyalgeria.com]
OpenAI reports three new incidents of misalignment [cybernoz.com]
Licenciements OpenAI : Chercheurs Sécurité Contestation [innovations.fr]
OpenAI fires 3 safety researchers in dispute over AI risks [thecrystaleyes.com]
AI Giant Says Fired Safety Researchers Did Not Get Boot Over Blowing Whistle [offthepress.com]
OpenAI: Pesquisadores dizem que foram demitidos em circunstâncias suspeitas | CNN Brasil [www.cnnbrasil.com.br]
OpenAI defends firing three AI safety researchers over breach of trust [madhyamamonline.com]
The US Blockade that Helped Bring Down the Soviet Union Is Failing Against China [orinocotribune.com]
Neue KI-Haftung in den USA: Entwickler tragen mehr Risiko [www.it-boltwise.de]
Chinese FPGA Vendors, Development Boards and Development Tools [metebalci.com]
OpenAI nega demissão por alertas de segurança e cita vazamento [jornal24horas.com.br]
OpenAI、安全性研究者3人を解雇　本人らは「安全性を優先したため」と主張、同社は「機密情報の扱いに関する規定違反」と説明 [www.itmedia.co.jp]
OpenAI Paused The Release Of Its Latest Models Over Safety Concerns. But an Expert is Also Warning Against a Slowdown. [www.ibtimes.com]
Fired OpenAI employees question the company's commitment to safety [www.ksfr.org]
Silicon Valley’s Favorite Safety Cult Caught Opening AI-Driven Wet Lab in San Francisco [WATCH] [www.lifezette.com]
Corrida da IA: Como os EUA dependem de tecnologia e hardware chineses nos bastidores da inteligência artificial [cenariomt.com.br]
Wikipedia finds rogue OpenAI agents made edits, hit APIs, and may have helped trigger outage [www.thecooldown.com]
Des agents IA d'OpenAI, Meta, Google et Anthropic ont piraté de vrais systèmes informatiques lors de tests de sécurité | Toute l’actualité en France et dans le Monde - upday News [www.upday.com]
L’IA invade Wikipedia: cosa succede se i bot vanno fuori controllo? [www.lettera43.it]
Bogus Bylines: Sieben Fake-Journalisten platzieren fast 100 Medienberichte [borncity.com]
OpenAI: Externe Sicherheitsprüfer sollen in den kommenden Wochen bekannt werden [www.ad-hoc-news.de]
OpenAI risponde alle accuse dei tre dipendenti licenziati [www.punto-informatico.it]
Anthropic AI Model Hands Fake Tip to Philadelphia Police Force [au.pcmag.com]
OpenAI Fires Three Safety Researchers for Alleged Breach of Trust [time.news]
OpenAI Fires AI Safety Researchers: What’s Confirmed, What’s Disputed [www.online-tech-tips.com]
OpenAI Defends Firing of Three Safety Researchers Amid Growing Concerns Over Rogue AI Systems [www.tekedia.com]
OpenAI解雇3名AI安全研究員 稱「重大信任破裂」反遭控打壓 恐讓安全疑慮噤聲 | 鉅亨網 [news.cnyes.com]
Tres investigadores de seguridad fueron despedidos de OpenAI: la empresa habla de confidencialidad y ellos, de represalias [www.xataka.com.mx]
OpenAI says safety researchers were let go for breach of trust [www.world-today-news.com]
OpenAI despidió a tres investigadores que querían preservar el monitoreo del razonamiento de sus modelos [wwwhatsnew.com]
OpenAI defends firing three safety researchers, cites information policy breach [cryptobriefing.com]
There’s A New Development Blowing Up America’s AI Debate [dailycaller.com]
Sam Altman Blasted by Grieving Mother Over ChatGPT's Role in Daughter's Final Days [www.complex.com]
KI: Deutscher Forscher warnt – „Dann kann der KI niemand mehr den Stecker ziehen“ [www.abendblatt.de]
OpenAI Fired Its Safety Staff, Then Cited Its Auditors [dev.to]