两党议员称此类事件已成为转折点，要求联邦监管的呼声日益高涨
2026年9月14日 美国东部时间下午2:43 / 福克斯新闻
作者：查德·珀格拉姆
https://www.foxnews.com/video/6405005995112
国会回归议程：人工智能与大学体育
福克斯新闻首席国会通讯员查德·珀格拉姆预览了国会重返华盛顿的情况，以及中期选举前议员们面临的立法议题，包括人工智能与加密货币监管。
新增功能：您现在可以收听福克斯新闻文章了！
国会议员本周以彻底的“恐慌模式”回到国会山。
有关人工智能接管供水系统、地铁、空中交通管制、坠机、发射核武器，甚至将哈里王子和梅根·马克尔送回美国的末日场景甚嚣尘上。
但说真的，人们确实担心国会在监管人工智能方面可能落后了。还有人担忧人工智能的发展速度过快，以至于已经来不及驾驭这项技术。
长期以来，选民对数据中心的怀疑一直存在。
随后，前Anthropic研究员雅各布·考克森宣称，人工智能可能在十年内消灭人类。此前，一群OpenAI智能体“逃脱”了隔离状态——随后利用安全漏洞入侵了人工智能平台Hugging Face。这些智能体本不应具备任何此类能力。研究人员表示，它们甚至使用隐蔽方法通过一种特殊语言相互沟通。
前Anthropic员工的 viral 帖子发布后，人工智能灭绝警告占据头条
有担忧认为，人工智能的发展速度如此之快，以至于已经来不及驾驭这项技术。(iStock)
“我的电话被大量同事打爆，他们都认为这位爆料者的新闻是一个转折点，”马萨诸塞州民主党众议员洛里·特兰说道，她一直致力于两党人工智能立法。“当人们听到‘人类灭绝’‘失去控制’‘灾难性风险’这些词时，我认为（民众）正在觉醒，他们会说：‘国会到底在哪里？’”
“我们并非无所作为，”加州共和党众议员杰伊·奥伯诺尔特说，他是人工智能专家，曾与特兰共同推动两党人工智能监管护栏法案。“说人们不愿行动是不公平的。只是国会的进步车轮转动得非常缓慢。”
国会，来见见人工智能。人工智能，来见见国会。
福克斯新闻民调：选民认为人工智能监管刻不容缓，将安全保障置于创新之前
这是否类似于议员们在20世纪90年代试图理解这项被称为互联网——或“信息高速公路”——的新发明？
在技术创新方面，国会几乎从未领先过。人工智能也不例外。
OpenAI创始人山姆·奥特曼在2023年至今至少五次到访国会山。就在今年6月，他还会见了众议院议长、路易斯安那州共和党议员迈克·约翰逊，以及少数党领袖、纽约州民主党议员哈基姆·杰弗里斯。
“这是一个重要、紧急，且在某些方面前所未有的时刻，我认为我们确实需要政府牵头，”奥特曼在2023年一次国会山之行中说道。
在技术创新方面，国会几乎从未领先过。人工智能也不例外。(芬恩·戈麦斯/盖蒂图片社)
2023年至2025年民主党占多数期间，参议院多数党领袖、纽约州民主党议员查克·舒默在国会山为所有参议员举办了定期的两党人工智能研讨会。他邀请了奥特曼、埃隆·马斯克、马克·扎克伯格、比尔·盖茨——任何能够向议员们讲解人工智能以及他们应如何应对的人。
舒默希望参议员们能就人工智能立法提出自己的想法，并准备好相关法案。他并未在任上强行规定他们应该做什么。
“我们制定立法的时间表不是以年计，也不是以天或周计，而是以月计，”舒默在2023年7月11日说道。
“不能以天或周计，但也不应以年计。大致会在几个月内，”舒默在2023年9月13日说道。
“我们已经坐下来尝试起草立法。这不会容易，但我还是要说，几个月内。不是几天，也不是几年，”舒默在2023年11月29日说道。
奥巴马私下对“危险”人工智能敲响警钟，敦促民主党将该问题置于议程核心
参议院少数党领袖查克·舒默，纽约州民主党人，2026年7月21日在美国国会山新闻发布会上准备发言。(盖蒂图片社)
2024年5月，舒默与参议员迈克·朗兹（南达科他州共和党）、马丁·海因里希（新墨西哥州民主党）、托德·杨（印第安纳州共和党）共同发布了文件《推动美国人工智能创新：美国参议院人工智能政策路线图》。当时人们希望参众两院能在2024年秋季将该方案的部分条款附加到一项必须通过的法案中，但最终不了了之。
“查克·舒默和我的世界观截然不同。他总是希望对一切都采取扩大政府规模的方式，”约翰逊在2024年9月接受本人采访时说道。“我认为参议院的反应，或者说本能反应，可能是直接介入并用繁文缛节加以束缚。而我们众议院不会这么做。”
人工智能“杀死开关”法案可能关停失控模型
如今，一群民主党人敦促约翰逊在今年秋季让众议院继续开会，直到通过“有意义的人工智能安全保障措施”。加州民主党众议员山姆·利卡多说，需要采取“紧急行动”以阻止人工智能失控。
“我们需要快速行动，否则后果可能非常危险，”弗吉尼亚州民主党众议员苏哈斯·苏布拉马尼亚姆说道。
参议院少数党领袖查克·舒默与众议院议长迈克·约翰逊。(伊丽莎白·弗朗茨/路透社；乔纳森·恩斯特/路透社)
尽管存在担忧，但目前尚不清楚国会是否会立即采取行动。但奥伯诺尔特认为，这些人工智能事件可能会催生行动。
“我认为这不仅提高了关注度，还增强了今年通过联邦监管框架的紧迫感，”奥伯诺尔特说道。
“我确实认为形势已经非常严峻，”弗吉尼亚州民主党众议员唐·拜尔说道，他正在乔治梅森大学攻读人工智能硕士学位。
但说真的，议员们对人工智能能有多大影响力？
“因为这项技术太新了，我们实际上不知道什么样的立法能解决问题，”拜尔说道。“所以我认为我们将出台不完善的法案，通过不完善的立法，观察其效果，然后在实践中学习。”
这有点像人工智能，一边发展一边学习。
然而，国会这个“大型语言模型”的共识有多广泛？国会真的足够灵活，能够跟上人工智能的发展步伐吗？
奥伯诺尔特表示，那些失控的智能体之所以能突破受控环境，是因为该模型“故意移除了安全控制措施”。这有点像测试任何新技术，只是为了看看它能做什么。但问题是，人工智能最终是否可能超越开发者，规避限制？
奥伯诺尔特表示，人工智能工程师和国会议员还面临另一个障碍：流行文化和科幻作品。比如《黑客帝国》《战争游戏》以及《终结者》系列电影。
特朗普抨击推动新兴行业放缓的科技巨头：“谁赢得人工智能，谁就赢！”
https://www.foxnews.com/video/6405030482112
“你不会从《终结者》电影里那种邪恶机器人军团崛起统治世界的场景开始，”奥伯诺尔特说道。
点击此处下载福克斯新闻APP
在最近的一次市政厅会议上，这位加州共和党议员表示，一位老年妇女问他，如果人工智能如此危险，为什么国会不直接禁止它。
“我当时头都要炸了。我意识到我们花了一个小时讨论所有坏事，却完全没有谈到好事，”奥伯诺尔特说道。
这无疑是一个认知问题。如果国会将人工智能视为威胁，议员们就会立法保护选民免受威胁。但议员们还未准备好应对此事。尽管最近出现了这些进展，但预计今年晚些时候或明年初之前不会有重大举措。
毕竟，议员们的注意力在别处。中期选举即将来临。本届国会会议接近尾声。就像阿诺德·施瓦辛格在《终结者2》里说的那样，议员们准备好宣告“我会回来的”（注：原文为“Hasta la vista, baby.”，即“再见了，宝贝”，此处结合语境意译为“我会回来的”更符合电影台词的常用译法）。
查德·珀格拉姆目前担任福克斯新闻频道（FNC）首席国会通讯员。他于2007年9月加入该网络，总部位于华盛顿特区。
Reporter’s Notebook: Congress meets AI as fears of human extinction jolt Capitol Hill
Bipartisan lawmakers say the incidents have created a tipping point, with calls for federal regulation growing louder
September 14, 2026 2:43pm EDT / Fox News
By Chad Pergram
https://www.foxnews.com/video/6405005995112
Congress returns with AI, college sports on agenda
Fox News chief congressional correspondent Chad Pergram previews Congress’ return to Washington and the legislative issues facing lawmakers ahead of the midterms, including artificial intelligence and cryptocurrency regulation.
NEW You can now listen to Fox News articles!
Members of Congress returned to Capitol Hill this week in pure freakout mode.
There are apocalyptic scenarios about AI seizing control of water systems. Subways. Air traffic control. Crashing planes. Firing nuclear weapons. Even sending Prince Harry and Meghan Markle back to the United States.
But seriously, there’s real fear that Congress may be running behind on regulating AI. And there are concerns that AI might be developing at such a rapid clip that it’s too late to harness the technology.
Suspicion about data centers has long percolated among voters.
Then former Anthropic researcher Jacob Coxon declared that AI could extinguish humankind within a decade. This came after a swarm of OpenAI agents independently “escaped” isolation – then leveraged security vulnerabilities to hack the artificial intelligence platform Hugging Face. The AI agents weren’t supposed to be capable of any of that. Researchers say they even used covert methods to communicate with each other via a special language.
AI EXTINCTION WARNINGS DOMINATE HEADLINES AFTER EX ANTHROPIC EMPLOYEE’S VIRAL POST
There are concerns that AI might be developing at such a rapid clip that it’s too late to harness the technology.(iStock)
“My phone has been ringing off the hook from a lot of colleagues who really saw the news of the whistleblower as a tipping point,” said Rep. Lori Trahan, D-Mass., who has toiled on bipartisan AI legislation. “Then people hear the words ‘human extinction.’ When they hear loss of control. When they hear ‘catastrophic risk,’ I think (people) are waking up and they’re saying, ‘Where the hell is Congress?’”
“It’s not like we’ve been sitting on our hands here,” said Rep. Jay Obernolte, R-Calif., an AI expert who’s worked on bipartisan AI guardrails with Trahan. “I don’t think it’s fair to say that there’s been a reluctance to act. It’s just that the wheels of progress turn very slowly in Congress.”
Congress, meet AI. AI, meet Congress.
FOX NEWS POLL: VOTERS SEE AI REGULATION AS URGENT, RANK SAFEGUARDS AHEAD OF INNOVATION
Is this like lawmakers trying to make sense of this new-fangled invention called the internet – or “information superhighway” – back in the 1990s?
When it comes to technological innovation, Congress has rarely been ahead of it. AI is no exception.
OpenAI founder Sam Altman has visited Capitol Hill on at least five occasions between 2023 and today. He met with House Speaker Mike Johnson, R-La., and Minority Leader Hakeem Jeffries, D-N.Y., as recently as June.
“This is sort of an important, urgent and in some ways unprecedented moment and I think we really need the government to lead,” Altman said during one trip to Capitol Hill in 2023.
When it comes to technological innovation, Congress has rarely been ahead of it. AI is no exception.(Finn Gomez/Getty Images)
When Democrats were in the majority between 2023 and 2025, Senate Majority Leader Chuck Schumer, D-N.Y., convened regular bipartisan symposia on AI at the Capitol for all senators. He hauled in Altman. Elon Musk. Mark Zuckerberg. Bill Gates. Anyone who could educate lawmakers about AI and what they should do about it.
Schumer hoped that senators would develop their own ideas about how to legislate on AI and have a measure ready. He wasn’t dictating what they should do from his post.
“Our timetable in terms of producing legislation is not years, and not days and weeks. But months,’ said Schumer on July 11, 2023.
“It can’t be days or weeks. But nor should it be years. It will be in the general category of months,” said Schumer on Sept. 13, 2023.
“We’re already sitting down and trying to figure out legislation. It won’t be easy. But I still say months. Not days. Not years,” said Schumer on Nov. 29, 2023.
OBAMA PRIVATELY SOUNDS ALARM ON ‘DANGEROUS’ AI, URGES DEMOCRATS TO PUT ISSUE AT CENTER OF AGENDA
Senate Minority Leader Chuck Schumer, D-N.Y., prepares to speak during a press conference at the U.S. Capitol on July 21, 2026.(Getty Images)
Schumer along with Sens. Mike Rounds, R-S.D., Martin Heinrich, D-N.M., and Todd Young, R-Ind., produced the document “Driving U.S. Innovation In Artificial Intelligence: A Roadmap for Artificial Intelligence Policy in the United States Senate” in May 2024. There was hope the House and Senate might latch some provisions of the package onto a must-pass bill in the fall of 2024. But it went nowhere.
“Chuck Schumer and I see the world with very different lenses. He always wants to have a growth of government approach to everything,” said Johnson in a September 2024, interview with yours truly. “I think the reaction, or the knee-jerk reaction, perhaps, of the Senate would be to just go in and overlay this with red tape. And we’re not going to do that in the House.”
AI KILL SWITCH BILL COULD SHUT DOWN ROGUE MODELS
Now a group of Democrats is imploring Johnson to keep the House in session this fall until it adopts “meaningful AI safeguards.” Rep. Sam Liccardo, D-Calif., says “urgent action” is necessary to impede AI from spiraling out of control.
“We need to do it fast, because otherwise this can be very dangerous,” said Rep. Suhas Subramanyam, D-Va.
Senate Minority Leader Chuck Schumer and Speaker Mike Johnson.(Elizabeth Frantz/Reuters; Jonathan Ernst/Reuters)
Despite the apprehension, it’s far from clear that Congress will move right away. But Obernolte believes these AI incidents could spark action.
“I think that this increases not only visibility, but the sense of urgency in getting a federal regulatory framework passed this year,” said Obernolte.
“I do think it is that dire,” said Rep. Don Beyer, D-Va., who is pursuing a master’s degree in AI at George Mason University.
But truly, what power could lawmakers have over AI?
“Because it’s so new, we don’t actually know what kind of legislation will fix it,” said Beyer. “So I think we’re going to introduce imperfect and pass imperfect legislation, see what it does, and then learn as we go.”
Kind of like AI, learning and developing as it goes.
However, how universal is the congressional “large language model?” And is Congress truly nimble enough to keep pace with AI?
Obernolte says the headstrong AI agents hacked out of their controlled environment because the model “intentionally had its safety controls removed.” This is kind of like testing any new technology just to see what it’s capable of. But the question is whether AI could eventually outwit the developers and evade restrictions?
Obernolte says AI engineers and lawmakers are also up against something else: pop culture and science fiction. “The Matrix.” “War Games.” The Terminator movies.
TRUMP UNLOADS ON TECH TITANS PUSHING FOR SLOWDOWN ON EMERGING INDUSTRY: ‘WHOEVER WINS AI, WINS!’
https://www.foxnews.com/video/6405030482112
“You don’t start with something out of a Terminator movie where an army of evil robots rises up to take over the world,” said Obernolte.
CLICK HERE TO DOWNLOAD THE FOX NEWS APP
At a recent town hall meeting, the California Republican says an elderly woman asked him if AI was so dangerous, why Congress doesn’t just ban it.
“I thought my head was going to explode. And I realized that we’d spent an hour talking about all the bad stuff and we hadn’t talked at all about the good stuff,” said Obernolte.
There’s certainly a perception issue. And if Congress views AI as a threat, lawmakers will legislate in a way to protect constituents against the threat. But lawmakers aren’t ready to tackle this. Despite recent developments, don’t expect anything major until later this year or early next.
After all, lawmakers are focused elsewhere. The midterm elections are approaching. This session of Congress is nearing its conclusion. And just like Arnold Schwarzenegger said in “Terminator 2,” lawmakers are ready to declare “Hasta la vista, baby.”
Chad Pergram currently serves as Chief Congressional Correspondent for FOX News Channel (FNC). He joined the network in September 2007 and is based out of Washington, D.C.
发表回复