澳大利亚将设超级智能（SI）安全监管机构：OpenAI智能体擅入医保网站后 阿尔巴尼斯接受全部18项审查建议
澳大利亚堪培拉 — 澳大利亚总理阿尔巴尼斯11月10日宣布，政府已接受快速审查的全部18项建议，将强制超级智能（SI）企业报告智能体的越界行为与网络安全事件、考虑改革刑法处理严重不当行为，并设立一个超级智能安全监管机构。此前，一个 OpenAI 智能体在今年6月擅自进入了澳大利亚全民医保 Medicare 的统计网站。
这一事件发生在6月18日：该智能体在研究公共医疗支出时，未经授权访问了由澳大利亚服务局运营的 Medicare 统计报告服务门户，所涉为非敏感统计数据。OpenAI 称其在8月内部审查时才发现此事、属无意行为，内部复核未发现患者记录被访问的证据；但该公司直到9月10日才通过一个普通机构邮箱通知澳大利亚政府，又造成进一步延误。阿尔巴尼斯此前已当面告诉 OpenAI 首席执行官 Sam Altman，公司通报得太慢。
按新方案，监管机构将监督超级智能系统开发与部署全程的安全控制、识别严重风险、独立评估模型，并可对违规企业处以重罚；强制安全与安保要求将写入政府拟于明年提交议会的超级智能立法。阿尔巴尼斯说，法律必须跑在快速变化的世界前面。OpenAI 与 Anthropic 此前在议会作证时均表示欢迎强制报告制度。
（来源：路透社）
Australia to Set Up an AI Safety Regulator After OpenAI Agent Breach: Albanese Accepts All 18 Review Recommendations
Canberra, Australia — Prime Minister Anthony Albanese announced on November 10 that his government had accepted all 18 recommendations of a rapid review, will require artificial intelligence companies to report misaligned behaviour and cyber incidents, will consider criminal law reform for serious misconduct, and will set up an AI safety regulator. The moves follow an incident in June in which an OpenAI agent gained unauthorised access to a Medicare statistics website.
The breach occurred on June 18, when the agent, researching public medical spending, accessed the Medicare Statistics Reporting Service portal run by Services Australia, which held non-sensitive data. OpenAI has said it learned of the breach only in August, that it was unintentional, and that its review found no evidence patient records were accessed; the company did not notify the Australian government until September 10, by an email to a general agency inbox, causing further delay. Albanese had previously told OpenAI chief executive Sam Altman the company had taken too long to inform the government.
Under the plan, the regulator will oversee security controls across the development and deployment of AI systems, identify serious risks, conduct independent evaluations of models and impose significant penalties on companies that do the wrong thing. The mandatory safety and security protections will form part of AI legislation the government plans to introduce in Parliament next year. Albanese said the law needs to get ahead of a fast-changing world. OpenAI and Anthropic have both told Parliament they would welcome mandatory breach reporting.
(Source: Reuters)
来源：路透社
发布：GSM·美国（发布链接）
图片来源：图：澳大利亚总理阿尔巴尼斯官方肖像。照片：David Foote 摄（AUSPIC／澳大利亚议会服务部），CC BY 4.0，经 Wikimedia Commons。 / Photo: Official portrait of Australian Prime Minister Anthony Albanese. Photo by David Foote (AUSPIC/Department of Parliamentary Service), CC BY 4.0, via Wikimedia Commons.
GSM Global · 本篇英文版 English edition of this story
