美国财政部长Scott Bessent指控中国AI公司Moonshot不当蒸馏Anthropic的Fable模型，并以此为由威胁对其进行制裁[1]。
与此同时，能源基础设施的优化也成为美国关注的焦点。纽约州电网在7月3日的高温期间从加拿大进口电力，以满足约9%的总电力需求[1]。支撑这一能源跨境供应的是今年5月开通的一条关键输电线路——魁北克至皇后区的地下输电线，全长339英里，是北美最长的地下输电线[1]。
此外，OpenAI在Hugging Face遭遇黑客攻击事件也引发关注。Colin Shea-Blymyer指出该事件展示了大语言模型网络操作的最高自主水平[1]，反映出当前AI系统自主性可能超出控制能力的风险。
The U.S. Treasury Department has threatened sanctions against Chinese artificial intelligence company Moonshot, accusing it of improperly distilling Anthropic's Fable model [1]. Treasury Secretary Scott Bessent made the allegation, marking an escalation in U.S. efforts to restrict technology transfer to Chinese AI firms [1].
In related developments concerning AI security, the OpenAI platform experienced a breach on Hugging Face that raised significant concerns about the autonomous capabilities of large language models [1]. According to Colin Shea-Blymyer, the incident demonstrated the highest level of autonomy in network operations achieved by large language models to date [1]. The breach underscores growing risks that AI systems may operate beyond human control parameters [1].
Separately, energy transmission infrastructure in North America continues to expand to meet demand pressures [1]. New York's electrical grid imported power from Canada to meet approximately 9 percent of total demand during a heat event on July 3, with supply routed through a 339-mile underground transmission line connecting Quebec to Queens that opened in May of this year [1]. This line represents North America's longest underground transmission corridor [1].
