据腾讯科技报道，人工智能初创公司OpenAI近日宣布了一项重大的治理变革，旨在提升其人工智能模型的安全性和保障措施。在这一变革中，OpenAI成立了“安全与保障委员会”，该委员会的成立是在公司负责安全领域的高管辞职、并有效解散其内部团队之后的举措。这一委员会的成立，标志着OpenAI对人工智能模型安全性的高度重视，同时也显示出公司对于加强内部治理结构、确保技术使用的透明度和责任性的决心。
安全与保障委员会的首要任务是评估和进一步发展OpenAI的流程和保障措施，这一任务将在未来的90天内进行。在评估结束后，委员会将与全体董事会成员分享其建议。随后，董事会将对这些建议进行审查，并在确保安全和保障的前提下，决定采纳哪些建议，并公开分享最新进展。这一系列举措表明，OpenAI正在积极采取措施，以应对人工智能领域日益增长的安全挑战，同时确保其技术的开发和应用能够符合最高的道德标准和法律要求。
与此同时，OpenAI还宣布开始训练一款全新的旗舰级人工智能模型，这一模型将超越并接替目前驱动其在线聊天机器人ChatGPT的GPT-4技术。这一举动标志着OpenAI在人工智能领域的持续创新和领先地位，同时也引发了业界对于新模型技术性能和应用潜力的广泛关注。
OpenAI的这一系列动作，不仅体现了公司对于人工智能技术安全性的重视，也展示了其在人工智能研究和开发领域的持续投入和创新精神。随着新模型的训练和未来可能的发布，OpenAI有望为人工智能领域带来更多的突破和创新，同时也为社会提供更加安全、可靠和负责任的人工智能技术。
英语如下：
News Title: “OpenAI Forms Security Committee, Launches Flagship Model Training, Responds to Governance Turmoil”
Keywords: OpenAI, Security, Model
Content: Tencent Technology reported that the AI startup, OpenAI, has recently announced significant governance changes aimed at enhancing the security and safeguards of its AI models. This initiative follows the resignation of the company’s head of security and the effective disbandment of its internal team. The establishment of the “Security and Assurance Committee” signifies OpenAI’s high regard for the security of its AI models and its determination to strengthen its internal governance structure, ensuring transparency and accountability in the use of technology.
The primary task of the Security and Assurance Committee is to evaluate and further develop OpenAI’s processes and safeguards, which will be conducted within the next 90 days. Upon completion of the assessment, the committee will share its recommendations with all board members. The board will then review these suggestions, with the intention of adopting those that ensure safety and safeguards, and will publicly share the latest developments. These measures demonstrate OpenAI’s proactive approach to addressing growing security challenges in the AI field, while ensuring that the development and application of its technology adhere to the highest ethical standards and legal requirements.
Concurrently, OpenAI has initiated the training of a new, flagship AI model that will surpass and replace the GPT-4 technology driving its online chatbot, ChatGPT. This move underscores OpenAI’s ongoing innovation and leadership in the AI field, attracting attention from the industry regarding the new model’s technical capabilities and potential applications.
OpenAI’s series of actions not only reflect the company’s commitment to AI technology security but also showcase its continuous investment and innovative spirit in AI research and development. As the new model is trained and potentially released in the future, OpenAI is poised to bring about more breakthroughs and innovations in the AI field, while also providing society with more secure, reliable, and responsible AI technology.
【来源】https://new.qq.com/rain/a/20240528A08IIZ00
Views: 2