Shocking Breach: Hacker Steals OpenAI Secrets, Sparking Alarming Fears of Chinese Espionage
Early last year, a hacker infiltrated OpenAI’s internal messaging system, extracting details about the company’s AI technologies from an employee forum. However, the hacker did not access the systems where OpenAI develops its AI. In an April 2023 meeting, OpenAI executives disclosed the breach to employees but opted not to alert the public, as no customer or partner data was compromised, and the hacker was not linked to any foreign government.
This incident has sparked internal concerns about potential espionage from foreign adversaries, particularly China. Leopold Aschenbrenner, a former OpenAI technical program manager, argued that the company wasn’t doing enough to secure its technologies from being stolen by foreign entities. Aschenbrenner, who was dismissed for allegedly leaking information, has criticized OpenAI’s security measures publicly.
OpenAI responded, stating that Aschenbrenner’s termination was not related to the concerns he raised. Despite generally low evidence of AI posing significant national security risks, some researchers and executives remain wary of future threats, including bioweapons or cyber-attacks.
Companies like OpenAI, Anthropic, and Google add security measures to their AI systems to prevent misuse. Meanwhile, federal officials and lawmakers are contemplating regulations to mitigate potential harms from AI technologies.
The global AI race continues, with China producing nearly half of the world’s top AI researchers. Some experts predict China could surpass the U.S. in AI capabilities. Clément Delangue of Hugging Face underscored the probability of China leading the field. National security leaders and researchers urge tighter controls on AI development to preempt future risks.
OpenAI has initiated steps to bolster security, including forming a Safety and Security Committee led by former NSA head Paul Nakasone. As the AI landscape evolves, the debate over its risks and ethical management persists.
Original Story https://www.nytimes.com/2024/07/04/technology/openai-hack.html
Category : Artificial Intelligence,Computers and the Internet,ChatGPT,Computer Security,Cyberattacks and Hackers,Open-Source Software,OpenAI Labs,Amodei, Daniela
Tags:
