内容来源：https://www.wired.com/story/uncanny-valley-podcast-inside-elon-musks-midterm-spending-spree/

内容总结：

本周，WIRED旗下播客节目Uncanny Valley由主持人Zoë Schiffer与Leah Feiger深入探讨了多条科技与政治交叉领域的重要议题。

在政治方面，埃隆·马斯克的America PAC正投入数百万美元，瞄准2026年中期选举中最关键的参议院选战。WIRED为此专门创建了实时追踪工具，持续监测该超级政治行动委员会的资金流向。数据显示，America PAC的支出策略与以往有所不同——其资金主要用于攻击民主党候选人，而非支持共和党人。每花一美元支持共和党候选人，就有超过两美元用于反对民主党候选人。在得克萨斯州，该组织已投入约900万美元攻击民主党人James Talarico，而用于推广Ken Paxton的资金仅约100万美元。俄亥俄州、艾奥瓦州等关键选战也呈现类似模式。这些选区的选择并非随机，它们恰恰是民主党夺回参议院控制权的关键路径，而一旦民主党掌控国会，马斯克及其关联企业可能面临更多调查和监管压力。

在科技思想领域，节目深入剖析了“有效利他主义”这一曾经小众的哲学理念如何走向主流，并成为人工智能安全辩论的核心。该理念主张人们应尽可能高效地利用时间、金钱和资源，去改善乃至拯救尽可能多的生命。由于许多AI研究者成长于这一哲学传统，他们将先进AI视为人类存续的重大风险，从而使有效利他主义与AI安全议题紧密交织。然而，特朗普政府对此持强烈批评态度，五角大楼高级官员将其称为“危险意识形态”，国防部官方社交媒体账号甚至呼吁公众选择“美国主义，而非有效利他主义”。节目还讨论了OpenAI的山姆·奥特曼与Anthropic的达里奥·阿莫代伊之间在这一问题上的分歧——前者代表“增长至上”的资本主义AI观，后者则更接近有效利他主义立场，强调AI安全优先。

此外，节目还关注了一起设计界的网络风波。Meta公司新AI助手Muse的标志设计师、自由职业设计师兼字体艺术家Jessica Hische在向订阅者分享该项目后，遭到了Threads平台上的大规模攻击。批评者认为她不应为AI技术效力，有人甚至发布她住宅的照片并对其财务状况进行揣测。Hische表示，近年来她的企业设计业务大幅萎缩，Meta提供的报酬约为12500美元，她需要用这笔钱来维持自己在奥克兰的两家小店——一家文具店和一家儿童艺术工作坊。尽管遭受猛烈抨击，她表示并不后悔公开这一项目，并认为大科技公司若因担心艺术家在社交媒体上批评它们而不再聘请独立创作者，最终受损的将是艺术家群体。

节目最后，两位主持人分享了各自的“WIRED/Tired”推荐。Leah Feiger推荐Netflix新一季《英国烘焙大赛》作为十月放松身心的舒适之选，同时提醒读者关注新闻之余也要适时放下手机。Zoë Schiffer则表示对当代文学中泛滥的第一人称现在时叙事感到厌倦，并推荐了作家George Saunders的短篇作品及其“第三人称腹语术”写作风格。

中文翻译：

本周的《怪诞谷》节目中，佐伊·希弗和莉亚·法伊格深入探讨了埃隆·马斯克的美国政治行动委员会如何斥资数百万美元介入2026年中期选举中最关键的一些参议院选战——以及其中大部分资金是如何用于打击民主党人、而非支持共和党人的。她们还剖析了有效利他主义——这个曾经小众的哲学流派如今正处于AI安全辩论的中心，以及为什么特朗普政府将其称为一种危险的意识形态。此外，她们还讨论了Meta公司AI助手Muse标志的设计师所遭遇的抵制风波。
请在信号奖中为我们投票！

• 节目类，最佳科技
• 节目类，最佳访谈或脱口秀
• 单集类，最佳纪录片
  本期节目提到的文章：
• 埃隆·马斯克的美国政治行动委员会正在中期选举中大举花钱。我们每天追踪
• 埃隆·马斯克的美国政治行动委员会正在悄悄资助十一月最关键的选举
• 花掉埃隆·马斯克的钱
• 她设计了Meta的新AI标志。然后仇恨来了
  你可以在Bluesky上关注佐伊·希弗（@zoeschiffer）和莉亚·法伊格（@leahfeiger）。给我们写信请发至[email protected]。
  如何收听
  你随时可以通过本页面的音频播放器收听本周的播客，但如果你想免费订阅以获取每一集，方法如下：
  如果你使用的是iPhone或iPad，打开名为“播客”的应用，或直接点击此链接。你也可以下载Overcast或Pocket Casts等应用，搜索“uncanny valley”。我们在Spotify上也有。
  文字记录
  注意：这是自动生成的文字记录，可能包含错误。
  莉亚·法伊格：大家好，我是莉亚。我们想告诉大家，《怪诞谷》入围了信号奖，不是一个或两个，而是三个奖项。评委决定谁获胜，但听众也可以通过“听众选择奖”投票来表达意见。你可以点击我们在节目笔记中附上的链接为我们投票。我们保证只需要一分钟，你的一票意义重大。好了，进入正题。
  佐伊·希弗：欢迎来到《连线》的《怪诞谷》。我是佐伊·希弗，特约编辑。
  莉亚·法伊格：我是莉亚·法伊格，政治与科学总监。本周我们将深入探讨埃隆·马斯克的美国政治行动委员会如何花费数百万美元来影响这个中期选举季中一些最关键的选战。《连线》创建了一个追踪器，实时跟踪美国政治行动委员会的支持动向。我们将带你了解为什么这些选战如此重要，以及马斯克如何向共和党人投入大量资金。
  佐伊·希弗：我们还将剖析那个曾经默默无闻、如今却似乎无处不在的哲学理念——从X上的帖子到《经济学人》封面——有效利他主义。我们将讨论这个意识形态到底是怎么回事，为什么许多AI研究者信奉它，以及它如何 actively 塑造着围绕AI安全的辩论。
  莉亚·法伊格：我们还会聊到为什么Meta公司Muse标志的设计师遭到了大量抨击。佐伊本周和她聊了聊，她有很多话要说。佐伊，这周只有你和我。我们正式把布莱恩推下了悬崖。
  佐伊·希弗：是的，我们把他推下去了。FaceTime聊天已经上线了。这周我们不会聊任何科技大会。
  莉亚·法伊格：绝对不会。但他下周就回来了。所以至少还有这个盼头。好了，佐伊，埃隆·马斯克。他又回来了。
  佐伊·希弗：他回来了。对某些人来说他从来没离开过。
  莉亚·法伊格：他确实从来没过离开，但他现在回来了，他的钱包也回来了。先帮大家回忆一下，马斯克是美国政治行动委员会的创始人和主要捐助者，这是一个超级政治行动委员会，2024年成立，旨在帮助唐纳德·特朗普总统入主白宫。他们成功了。特朗普已经就职。该委员会为此花费了超过2.6亿美元。这次针对十一月的期中选举，该委员会的做法略有不同。《连线》创建了一个实时追踪器来监测他们的捐款。你可以在《连线》网站上查看。我们在制作过程中学到了很多。基本上，该委员会正在向选举中最关键的选战捐款。他们显然没有聚焦总统大选。今年没有总统大选。但他们全力投入了全国最关键的参议院选战，这些选战将真正决定明年国会的构成。该委员会主要是砸钱反对民主党人，而不是支持共和党人，这非常有意思。美国政治行动委员会每花一美元支持共和党候选人，就有超过两美元用于反对民主党人。在某些参议院选战中，这个比例更加极端，对吧？在德克萨斯州——马斯克旗下多家公司的所在地——该组织花了大量资金攻击民主党人詹姆斯·塔拉里科。大约900万美元。而用于推广肯·帕克斯顿的只有大约100万美元。俄亥俄州参议院选战、艾奥瓦州，几乎到处都是类似的数据。而这些选战并非随机选择。它们代表了民主党今年十一月夺取国会控制权的路径，也可能让共和党人——包括马斯克——面临大量调查和监管压力。所以这在很多方面都至关重要。我非常期待和你聊这个话题。你觉得那个追踪器怎么样？你有机会试用一下吗？
  佐伊·希弗：我确实有机会试用了一下。我印象非常深刻，尤其是因为你们为互联网编写了一个应用和网页，说实话，这是开创性的。
  莉亚·法伊格：我一点编程功劳都不能认领。这全归功于我们了不起的调查团队，他们真的打了一个全垒打。他们确实做到了。
  佐伊·希弗：好吧，玩笑归玩笑。不，我觉得这非常重要。我的意思是，就像你说的，美国政治行动委员会的总支出比之前少，但目标性强得多。他们非常有针对性。这是因为赌注非常高。我觉得我们应该先退一步聊聊这个。因为国会确实有可能翻盘，民主党有可能夺回控制权，而马斯克和他的同僚们似乎对此非常非常重视。
  莉亚·法伊格：哦，他们当然应该重视。民主党人已经在告诉我们——我们所有的民主党消息源，无论在众议院还是参议院方面——都在虎视眈眈地谈论着一旦国会到手就要发出的所有调查和传票。而目前民调对他们来说相当有利。这很大程度上是——如果民主党不搞砸的话——他们很有机会实现这一目标。我只能想象马斯克和类似的一群人现在大概在抓狂。
  佐伊·希弗：我是公开表过态的，我再重复一遍，我认为共和党人会表现得很好的。我觉得我们之前有过这样的对话，我对民调持怀疑态度。不过，你是政治总监，所以我最终听你的。话虽如此，我认为马斯克确实非常非常认真地对待调查和监管的可能性。这是他最终决定全力支持唐纳德·特朗普的主要原因之一。他非常公开地说过，如果卡玛拉·哈里斯在上个选举周期赢得总统大选，整个国家都会变得像加州一样，而据说是那可能发生的最糟糕的事情。作为来自加州的人，我不同意，但这不是重点。在这里建东西确实很难。这一点我承认。话虽如此，我很难想象一个真正严厉的监管会落到埃隆·马斯克帝国头上的世界。我认为部分原因是SpaceX，以及政府对星链、对SpaceX技术、对他们将东西送入太空的能力有多么依赖。SpaceX现在是代表政府做这些事的公司。所以虽然我确实认为传票之类的事情可能发生，但它最终会影响埃隆·马斯克的帝国吗？坦率地说，我不确定我们现在是否有工具来让这样一个人承担责任。
  莉亚·法伊格：我觉得这是一个非常非常好的观点，说实话。而且显然有一个明确的担忧，就是某个类似莉娜·汗的人物可能会出现并制造混乱。但就像你说的，那种混乱将是文书工作。不，我听到你了。我听到你了。
  佐伊·希弗：我是说，想想看。他唯一一次真正被追究责任是因为他说他有钱将特斯拉私有化。他已经获得了将特斯拉私有化的资金保障。结果是什么？他被罚款2000万美元？埃隆·马斯克身价多少？最终，即使这个数字让他心疼，他不想付，那也不过是九牛一毛。
  莉亚·法伊格：不，他现在是万亿富翁了，反正是。而且说到这一点，这不是超级政治行动委员会或大资金政治第一次引发一些争议。我们在2024年有过一次大讨论，关于埃隆·马斯克往墙上扔钱有多疯狂，现在他又通过美国政治行动委员会扔钱了，这现在感觉像是一种模式，而且感觉其他科技创始人及其影响力和他们的政治行动委员会在这个政府中已经过于突出了。硅谷上上下下各种团体的政治行动委员会都深度参与了这次选举周期。我很想知道你听说了什么，关于人们如何看待这个中期选举周期，以及他们如何更多地考虑未来。
  佐伊·希弗：是的。我的意思是，硅谷不是铁板一块。你听到很多来自房间里声音最大的人的意见，而其中很多声音是全力支持特朗普总统和共和党的人。我认为这些人在你说的某些担忧上与埃隆·马斯克一致——会有调查，如果民主党掌权，会减缓增长，他们将无法在AI和其他领域领跑。所以他们需要全力投入，保护他们希望看到的未来。当然，也有很多普通员工持有更自由的价值观，他们比较低调，但他们在为不同的事业做贡献。现在硅谷的对话很有意思，因为有很多新晋百万富翁，在AI研究领域工作过、现在有很多钱的人，他们正在积极讨论现在部署资金、分配资本的最佳方式是什么？把钱放在哪里才能产生最大的效果？是慈善？是政治？如果是政治，我们应该捐给哪里？目前没有一个统一的定论，但这绝对是一个话题。莉亚，你和我们的听众都知道，关于AI安全的辩论最近无处不在。它已经在很大程度上突破了圈层。在这场辩论中，有一个核心哲学反复出现，并且开始成为关注的中心——有效利他主义。上个月，AI研究员雅各布·考克森大声表达了对有效利他主义核心关切之一的支持——先进AI可能毁灭人类。这引发了一波头条新闻和特朗普政府相当激烈的回应。五角大楼高级官员将有效利他主义标记为“危险的意识形态”，国防部的一个官方社交媒体账号恳请人们选择“美国主义，而非有效利他主义”。所以我们确实看到了这些分界线。但莉亚，先退一步说，你是什么时候开始知道的——因为你不在硅谷，你在纽约——你是什么时候了解到有效利他主义的？
  莉亚·法伊格：神奇的是，我实际上可以精确地定位到，大概16个月前，而且是从你那里知道的。我们在通电话，你说：“哦对对对。这个人全力支持有效利他主义。”我说：“我完全不知道那是什么意思。我以前从来没听说过。我从来没听说过这种信仰。”你真的应该向我们的听众描述一下，但它确实帮助我理解了硅谷在AI问题上存在分裂。我之前不知道这一点。我以为人们就是：“是的，我们为这些公司工作。我们准备好了。”然后你带我了解了Anthropic里的有效利他主义者和OpenAI里不同的一方，我相信那是幼儿园版本，但真的非常非常有帮助。所以请开始吧。
  佐伊·希弗：所以在最基本的层面上，有效利他主义是一种信仰体系，核心思想是人们应该尽可能高效地利用自己的时间、金钱和资源，来改善甚至拯救尽可能多的人和动物的生命。它的核心从根本上并不特指AI。它真正关注的是我们如何赚尽可能多的钱，然后以某种方式部署这些钱来帮助尽可能多的人或拯救尽可能多的人。有效利他主义的这些概念或理念是由像LessWrong这样的博客推广开来的，这些博客很小众，但在硅谷，它在2000年代初确实有真正的影响力。牛津的研究人员在讨论它。AI之所以成为其中一部分，是因为很多AI研究者是在这种哲学中成长起来的，他们将AI视为真正的存在性风险。如果你关心人类的福祉，然后你开始大量接触一种你认为也许没有巨大风险、但仍然有实质性风险会杀死所有人的技术，那么这就成了核心关切。不过我想指出一件事，虽然我远不是要为有效利他主义者摇旗呐喊，但像任何运动一样，它是一个光谱。有的人是真正的狂热分子。当你听到他们的观点以及他们为了拯救最多数量的人而愿意做出的取舍时，那种功利主义信仰体系听起来完全疯狂。但我和很多自认为是有效利他主义者的人聊过，他们完全正常。我本来想说理性，但在这个语境下这个词太有争议了。但他们在非常深入地思考这个问题。他们听起来不像宗教狂热分子。他们看起来不像邪教成员。而且并不是所有这些人都聚在一起讨论。这是一个运动，它很广泛。有不同的群体。
  莉亚·法伊格：当你把它应用到AI上时，我觉得有一个吸引人的方面，你是在说，看，我们只是在说我们应该关注这项技术将走向何方以及它可能对人产生什么影响。这是底线，对吧？这是一个相当不错的总结吗？
  佐伊·希弗：百分之百。如果你认为你正在研究的一项技术可能有10%的概率杀死所有人，理性的做法是辞职并警告所有人。所以从这个意义上说，雅各布·考克森所做的并不完全令人震惊。但看到这个完全进入主流是非常有意思的。我们看到了真正关心AI安全的AI研究者上了塔克·卡尔森、乔·罗根的节目。再说一次，有效利他主义者和AI安全研究者的维恩图不是完美的圆，但有相当程度的重叠。
  莉亚·法伊格：好，那萨姆·奥尔特曼、OpenAI、达里奥·阿莫代伊、Anthropic。达里奥谈过AI对世界的风险。那他算是EA类别吗？
  佐伊·希弗：这是一个非常非常有争议的问题。萨姆·奥尔特曼，OpenAI的CEO，和达里奥·阿莫代伊合作过一段时间。他们都在OpenAI。然后达里奥辞职了，和一小群OpenAI研究员一起创办了一家新公司，专门试图安全地构建AI以造福全人类等等。因为他们觉得在OpenAI做不到这一点。所以那里你就看到了一个明显的分裂——这不是他们自我认同的方式，但这是人们给他们分类的方式。萨姆·奥尔特曼某种程度上代表了不惜一切代价增长的人工智能资本主义观点。达里奥更多地代表了有效利他主义——让我们安全地做。我们必须是赢家，因为只有我们理解其中真正的风险。达里奥在最近几个月和几年里说过，我相信，他不认同自己是有效利他主义者。话虽如此，很多信念确实重叠。我觉得和Anthropic的研究人员交谈，你会感觉到他们中更多人似乎属于有效利他主义哲学。
  莉亚·法伊格：有专门的Slack频道吗？有聚会吗？我很好奇，既然这已经进入了主流词汇，我很想知道除了五角大楼说“要亲美，说不”之外，这会如何继续发展。
  佐伊·希弗：是的。我的意思是，是的，有群聊。我确信有Slack频道。伯克利有一个叫Lighthaven的地方，有点像会议中心/合租社区的合作空间，会举办活动。我上次去的时候，有人提出可以睡他们的双层床。我礼貌地永远拒绝了。不了，谢谢。但总之，有意思的是，随着这越来越主流，就像你说的，特朗普政府出来了，他们真的把有效利他主义当作一个笼统的标签，说这些人是近乎宗教狂热分子，他们反增长，反AI进步，因此反美国。当然这里有更多的细微差别。我认为很多人只是在非常认真地对待AI安全。但现在你看到有效利他主义变成了一个贬义标签，意思是我们不能也不应该认真对待这些人。
  莉亚·法伊格：为什么这有争议？你说的听起来没那么糟。我们只是在试图拯救人类？
  佐伊·希弗：是的。是的。我认为一个原因是这一切都关乎取舍。当推向极端时，你可以看到如果你的目标是最大化善或最小化伤害，你会愿意承担一定量的伤害，或者参与其中，我想，以拯救更多的人。我认为人们——某种程度上合理地说——有点被吓到了，因为为了消除更大伤害发生的可能性，有效利他主义者愿意在当下做一些有伤害性的事情，因为他们认为他们在保护人们免受以后可能发生的事情。我认为另一部分——这绝对是我在思考的——是如果你认为你是唯一认真对待AI安全的公司，那么你“赢得”AI竞赛就非常重要。如果你认为赌注是人类级别的灭绝，那你有什么不会去做来赢得胜利呢？
  莉亚·法伊格：我有点觉得参与这场对话的每个人都需要去欧柏林学院报一门哲学入门课。这可能是我现在的感受。
  佐伊·希弗：我觉得欧柏林就是让我们陷入这个困境的原因，也许吧。特别是欧柏林的哲学。
  莉亚·法伊格：102。哲学102。这是下一章。既然你已经从斯拉沃伊·齐泽克那里毕业了，接下来是什么？
  佐伊·希弗：天哪。说回正题，距离Meta发布其新AI助手Muse已经过去几周了，说实话，这对公司来说是个意外的爆款。它仍然在应用排行榜上名列前茅。但我想和你聊这个的原因，是几周前在湾区发生了一件非常有意思的事。设计师——一位自由职业设计师和字体艺术家——杰西卡·希舍，被Meta聘请帮助设计Muse标志，她告诉她的新闻通讯订阅者：“嘿，我做了这个项目。”事实上，我觉得新闻通讯的标题是“新作品！（别生我的气）”。她多少知道会遭到一些抨击。但她基本上说：“看，Meta聘请我做这个项目。最终，我认为一个大公司应该因为聘请独立艺术家而不是直接生成一个新标志而受到赞扬。设计过程是这样的。希望你们喜欢。”大概就是这个意思。她的一个新闻通讯订阅者把它发到了Threads上。你忘了Threads还存在。在设计圈，Threads还是很有存在感的。
  莉亚·法伊格：我确实忘了。我完全忘了。
  佐伊·希弗：然后人们就炸了。有人说对她感到厌恶、失望，说从她还在艺术学校时就关注她，再也无法支持她做的任何事情。有人发布她家的照片，猜测她的财务状况。她觉得整个Threads就像一群暴民攻击了她三天，就因为她出来说自己做了这份工作。是的。
  莉亚·法伊格：那真的太激烈了。
  佐伊·希弗：是的，确实很多。和她聊天非常有意思，因为她对AI其实有复杂的感受。她说她丈夫在这个领域工作过，他们实际上有一些分歧，因为他真的对这项技术很兴奋。如果你读了那篇文章，说实话非常有“AI妻子”的感觉。而她说：“不，这对商业艺术可能非常不利。”那些为了支持自己收入较低的艺术创作而偶尔为公司工作的艺术家——就像她一样——这可能是有害的。事实上，这就是杰西卡·希舍所做的。她在奥克兰有两家店。非常可爱。一家有点像文具店。另一家是她举办儿童艺术工作坊和课程等的空间。她基本上完全靠接企业工作来资助这些店。但她说过去几年，她的企业工作真的枯竭了，她收到的咨询不像以前那么多了。所以当Meta来找她时，她说：“我甚至没觉得有什么矛盾的。我想，天哪，我想给我的一个员工提供健康保险。我需要在每次Gusto通知我要发工资时不再那么紧张。是的，我接。”她拿到了报酬。猜猜重新设计Muse标志你能拿多少？
  莉亚·法伊格：我的意思是，那可是Meta。你应该拿到五十万美元的股权。
  佐伊·希弗：是的，你会这么想。不，她拿了大约12500美元。
  莉亚·法伊格：不会吧。不会吧。这是最糟糕的部分。这实际上是整个故事中最糟糕的部分。
  佐伊·希弗：我知道。是的。她说：“我觉得两天的活拿这个钱还挺不错的。”因为那是一个冲刺项目。她一开始只是在设计那个M，后来做了Muse的完整字体，所以多拿了一些。但是的，也没那么多。再说一次，她用这些钱来支持她在奥克兰的小生意。但她说Threads上有一大群人似乎是真正的AI纯粹主义者，他们从根本上认为：“在这项技术上工作就是没有任何借口。”他们真的把她决定做这份工作看作是一种支持的表态。
  莉亚·法伊格：我心情复杂。我不知道。去找公司算账，别找艺术家，这大概是我的感受。但我理解这不像在社交媒体应用上直接骂某个人那样有料或刺激。
  佐伊·希弗：一个由Meta拥有的社交媒体应用，说清楚一下。
  莉亚·法伊格：这一点非常重要。是的。她后悔吗？
  佐伊·希弗：有意思。我问了她这个问题，她说最终不后悔。她权衡了：“我后悔谈论这件事吗？”但她说最终她觉得通过赞扬一家聘请了独立艺术家的公司，她为艺术家做了件好事。她说：“这些大型科技公司如果觉得艺术家之后会在社交媒体上抨击他们，就不会聘请艺术家。最终，这些公司存在，可以与他们建立某种共生关系。”我个人对此有同感。
  莉亚·法伊格：如果我不至少提到或想到《硅谷》这部剧，那就不是《怪诞谷》的一集。我忍不住想到剧中那个为加文·贝尔森设计盒子的艺术家，以及那需要付出什么。我记得看的时候想，你为什么要让自己经历这些？但每个人都需要薪水。那很痛苦。哦，哇。这是个复杂的问题。听起来有点难受。
  佐伊·希弗：广告之后，我们将分享本周的《连线》/疲劳之选。别走开。
  ［广告］
  现在是我们的《连线》/疲劳之选环节。新潮酷炫的就是《连线》。我们完全过时的就是疲劳。莉亚，你这周有什么要分享的？
  莉亚·法伊格：好。我想先说每个人都应该订阅Wired.com，每个人都应该继续读新闻。就这样。没有“但是”。
  佐伊·希弗：很好。
  莉亚·法伊格：但我想补充的是，这很累人。这是一个非常漫长的选举周期。我们刚刚终于结束了初选季。我们现在进入了中期选举季。每天都有来自伊朗、中国和俄罗斯瘟疫的新闻。有时候真的可以停下来不刷手机，关掉新闻。我真的想提醒人们这一点。这就引出了我的《连线》推荐，可以帮你做到这一点。我不是被付费说这个的，虽然我有点希望是。但Netflix上有新一季的《英国烘焙大赛》，我真的真的很喜欢。就是舒适温馨，非常适合十月，强烈推荐。所以读几篇《连线》的文章，对世界局势感到紧张，看一集，然后回去做些其他有趣的事情。佐伊，你的呢？
  佐伊·希弗：好，我喜欢这个。在我们开始之前，你觉得你的平均屏幕使用时间是多少？相关的。这是相关的。
  莉亚·法伊格：哦。我会说大概五到六个小时。
  佐伊·希弗：好。上周我是14个小时。
  莉亚·法伊格：哇。
  佐伊·希弗：我一直在看手机。我知道。是不是很可怕？我的心理健康糟透了，最后我想，这有关联吗？我觉得自己好傻。
  莉亚·法伊格：哇。
  佐伊·希弗：是的。是的，我知道。我知道。真的太病态了。
  莉亚·法伊格：那太多了。我为你担心。
  佐伊·希弗：这周我降到了七个小时。我开始频繁查看，所以查看屏幕时间反而增加了屏幕时间。好。我的疲劳/《连线》是，我觉得我读的每一本现代书——这是一个粗暴的概括——第一人称现在时无处不在。以前的书不是这样写的，现在你打开一本普通的文学小说，就是这种写法，真的很烦。乔治·桑德斯，伟大的作家、老师等等，谈过一种写法，我想他称之为“第三人称腹语术”。你在人物的脑子里，但仍然是第三人称。所以叙述听起来像这个人在自言自语。我昨晚读了他为《纽约客》写的一个短篇，感觉像这种风格，我想叫《胜利之圈》。他真的是一个真正的艺术家。我非常尊重他。我真的很喜欢他关于创作过程和作为一个人存在于世所说和所写的一切。
  莉亚·法伊格：我喜欢这个。我要去读。很棒的推荐。你要去看《英国烘焙大赛》吗？
  佐伊·希弗：不，我不会，因为我还是没有电视。但我觉得总有一天我会回到那个状态，那会是一个让人感觉很好的节目。这就是我们今天的节目。我们会在节目笔记中链接我们谈到的所有文章。阿德里亚娜·塔皮亚制作了本期节目。普兰·班迪负责混音，他也是我们纽约演播室的工程师。马特·吉尔斯和丹尼尔·罗曼做了事实核查。凯特·奥斯本是我们的执行制片人，凯蒂·德拉蒙德是《连线》的全球编辑总监。
  评论
  回到顶部

英文来源：

This week on Uncanny Valley, Zoë Schiffer and Leah Feiger dive into how Elon Musk’s America PAC is spending millions on the most consequential Senate races of the 2026 midterms—and how most of that money goes to opposing Democrats rather than supporting Republicans. They also break down effective altruism, the once-niche philosophy now at the center of the AI safety debate, and why the Trump administration has called it a dangerous ideology. Plus, they discuss the backlash against the graphic designer who created the logo for Meta’s AI assistant, Muse.
Vote for us in the Signal Awards!

• Shows, Best in Technology
• Shows, Best in Interview or Talk Show
• Individual Episode, Best in Documentary
  Articles mentioned in this episode:
• Elon Musk’s America PAC Is Spending Big on the Midterms. We’re Tracking It Daily
• Elon Musk’s America PAC Is Quietly Funding November’s Most Consequential Elections
• Spend Elon Musk’s Money
• She Designed Meta’s New AI Logo. Then Came the Hate
  You can follow Zoë Schiffer on Bluesky at @zoeschiffer and Leah Feiger on Bluesky at @leahfeiger. Write to us at [email protected].
  How to Listen
  You can always listen to this week's podcast through the audio player on this page, but if you want to subscribe for free to get every episode, here's how:
  If you're on an iPhone or iPad, open the app called Podcasts, or just tap this link. You can also download an app like Overcast or Pocket Casts and search for “uncanny valley.” We’re on Spotify too.
  Transcript
  Note: This is an automated transcript, which may contain errors.
  Leah Feiger: Hey everyone, this is Leah. We wanted to share with you that Uncanny Valley is a finalist for not one or two, but three Signal Awards. The judges pick who wins, but listeners also have a say with the listener's choice vote. You can go to the links we included in the show notes and vote for us. We promise it takes just a minute and your vote goes a long way. OK, onto the show.
  Zoë Schiffer: Welcome to WIRED’s Uncanny Valley. I'm Zoë Schiffer, contributing editor.
  Leah Feiger: And I'm Leah Feiger, director of politics and science. This week we're doing a deep dive into how Elon Musk's America PAC has spent millions of dollars to influence some of the most consequential races this midterm season. WIRED has created a tracker to follow America PAC's support in real time. We're going to take you through why these races matter and how Musk is throwing Republicans all of this cash.
  Zoë Schiffer: We're also going to break down the once obscure philosophy that now seems to be everywhere from X threads to The Economist cover, effective altruism. We'll talk about what this ideology is actually all about, why many AI researchers subscribe to it, and how it has been actively shaping the debate around AI safety.
  Leah Feiger: And we're going to get into why the designer of Meta's Muse logo has been receiving a ton of backlash. Zoë spoke with her all about it this week and she had a lot to say. Zoë, it's just me and you this week. We've officially pushed Brian off a cliff.
  Zoë Schiffer: Yeah, we have. The FaceTime chat has gone live. We will not be talking about any tech conferences this week.
  Leah Feiger: Absolutely not. But he'll be back next week. So at the very least, there's that. OK, Zoë, Elon Musk. He's back.
  Zoë Schiffer: He's back. He never left for some.
  Leah Feiger: He never really left, but he is back and so is his wallet. To refresh everyone's memories a little bit, Musk is the founder and main donor of America PAC, a super PAC that was started in 2024 to help elect President Donald Trump to the White House. They succeeded. Trump is in office. The PAC spent more than $260 million to make that happen. This time around for November's midterms, the PAC has been going about it a little bit differently. WIRED created a live tracker to monitor their donations in real time. You can check it out on the WIRED website. And we learned a lot putting this together. Basically, the PAC is donating to the election's most consequential races. They are not focusing obviously on the presidential race. There is no presidential race this year. But they are all in on the most critical Senate races in the country that are really going to decide the makeup of Congress next year. The PAC is primarily pumping money against Democrats as opposed to for Republicans, which is really interesting. For every dollar that America PAC has spent supporting a Republican candidate, more than two have gone to oppose a Democrat. And that only becomes more extreme in some of these Senate races, right? In Texas where several Musk companies are based, the group has spent tons and tons of money attacking Democrat James Talarico. We're talking like $9 million. Whereas promoting Ken Paxton, around the one million mark. Similar numbers can be found in the Ohio Senate race, in Iowa, really up and down the board. And these races are not random. They represent the path for Democrats to take control of the chamber this November and could open Republicans, Musk included, up to a lot of investigations and regulatory pressure. So this is critical for so many reasons. I'm so excited to talk with you about it. What did you think of the tracker? Did you get a chance to play around with it?
  Zoë Schiffer: I did get a chance to play around with it. I was really impressed, not least because you coding something for the internet, an app and web, honestly, groundbreaking.
  Leah Feiger: I can take zero, zero coding credit. This all goes to our amazing investigations team who really knocked it out of the park. They did.
  Zoë Schiffer: Well, jokes aside. No, I think it's really important. I mean, like you said, the total spending on America PAC's part is less than what they spent previously, but it's much more targeted. They're being really specific. And that's because the stakes are really high. I feel like we should just back up for a second and talk about that. Because there is this chance that Congress flips, that Democrats take control, and it seems like Musk and his associates are taking that very, very seriously.
  Leah Feiger: Oh, as they should. Democrats are already telling us, all of our sourcing in Democrats, both on House and Senate side are hungrily talking about all of the investigations and subpoenas that they're going to release the minute that the chamber is theirs. And right now, the polls are looking pretty good for them. This is very much a, if Dems don't mess this up, they have a very good chance of making this happen. I can only imagine that Musk and a similar coterie of fellows are kind of freaking out right now.
  Zoë Schiffer: I mean, I am on the record and I'm going to say it again, I think Republicans are going to do just fine. I feel like we've had this conversation before, and I'm sus on the polling. However, you are the politics director, so ultimately I defer to you. That said, I think it is 100 percent true that Musk takes the possibility of investigations and regulations really, really seriously. This was a primary reason why he ultimately decided to go all in on Donald Trump. He said very publicly that if Kamala Harris won the presidential election during the last cycle, the whole country was going to look like California, which was the worst thing that could happen to it, apparently. As someone from California, I disagree, but neither here nor there. It is really hard to build things here. I'll give you that. That said, I find it very hard to imagine a world where really harsh regulations actually do come down on the Elon Musk empire. I think part of the reason for that is SpaceX, and just how dependent the government is on Starlink, on SpaceX technology, on their ability to get stuff into space. SpaceX is the company doing that on behalf of the government now. And so while I do think subpoenas, all of this stuff could happen, will it impact Elon Musk's empire ultimately? I don't know if we have the tools to hold someone like that accountable at this point, frankly.
  Leah Feiger: I think that's a really, really good point, honestly. And obviously there's a clear fear that a Lina Khan-esque figure could come in and wreak havoc. But to your point, that havoc will be paperwork. No, I hear you. I hear you.
  Zoë Schiffer: I mean, think about it. The one time he was held accountable in a real way was for saying that he had money to take Tesla private. He had funding secured to take Tesla private. What was the outcome of that? He was fined $20 million? How much is Elon Musk worth? Ultimately, even if that number hurts, he doesn't want to pay it, it's kind of a drop in the bucket.
  Leah Feiger: No, he's a trillionaire now, again, I guess. And I guess to that point, this is not the first time that super PACs or big money in politics has stirred up a little bit of controversy. We had this big conversation in 2024 about how crazy it was that Elon Musk was just throwing money at the wall, and now he's throwing money again via America PAC, which now feels like a pattern, and it feels like other tech founders and their influence and their PACs are already outsized in this administration. You have PACs from all sorts of groups up and down Silicon Valley that are very involved this election cycle. I'm really interested in what you're hearing about how folks are thinking about this midterm cycle and how they're kind of thinking more future.
  Zoë Schiffer: Yeah. I mean, Silicon Valley is not a monolith. You hear a lot from the really loud voices in the room, and a lot of those voices are people who have gone all in on supporting President Trump and the Republican Party. And I think those people kind of align with Elon Musk in some of the fears that you have said, that there will be investigations, that if Democrats take control, it'll slow growth, that they won't be able to lead the race in AI and these other areas. And so they need to go all in and protect that future that they want to see happen. Of course, there are a lot of rank-and-file employees who have more liberal values, and they are staying a little more quiet, but they are contributing to different causes. It's an interesting conversation in the Valley right now because there are a lot of newly minted millionaires, people who worked in AI research and now have a ton of money, and they are having active conversations on what is the best way to deploy, to allocate capital right now? Where do we put the money to have the biggest effect? Is it in philanthropy? Is it in politics? If it's in politics, where should we contribute? There's not one defining take right now, but it is definitely a topic of conversation. Leah, as you and our listeners both know, the debate over AI safety has been everywhere lately. It has escaped containment in a big way. And in that debate, there is one central philosophy that keeps coming up again and again, and it's starting to become kind of the center of attention, effective altruism. Last month, AI researcher Jacob Coxon loudly voiced support for one of effective altruism's central concerns, that advanced AI could destroy humanity. This kicked off a wave of headlines and a pretty intense response from the Trump administration. Senior Pentagon officials labeled effective altruism as a, quote, unquote, "dangerous ideology" and an official Department of Defense social media account implored people to choose, quote, "Americanism, not effective altruism." So we're really seeing these dividing lines. But Leah, just to back up a little bit, when did you become aware, because you're not in Silicon Valley, you're in New York, when did you learn about effective altruism?
  Leah Feiger: What's amazing is I can actually pinpoint it to, I would say, maybe 16 months ago, and it was from you. We were on a phone call, and you were like, "Oh yeah, yeah, yeah. And this person is all in on effective altruism." And I was like, "I have no idea what that means. I have never heard of this before. I have never heard of this belief." And you should really describe it to our listeners, but it really, really helped me understand that there was a split in Silicon Valley over AI. I wasn't aware of that before. I just though that people were like, "Yep, we work for these companies. We're ready to go." And you took me through the effective altruists in Anthropic and different side in OpenAI, and it was, I'm sure, like the kindergarten version, but it was really, really helpful. So please take it away.
  Zoë Schiffer: So at a very basic level, effective altruism is a belief system built around the idea that people should use their time, money, and resources as efficiently as possible to improve or even save the lives of as many people and animals as they possibly can. At its core, it doesn't fundamentally have anything to do with AI specifically. It really is how can we make as much money as possible and then deploy it in such a way to help as many people as possible or save as many people as possible? This concept or these ideas of effective altruism were popularized by blogs like LessWrong, which is fringe, but in Silicon Valley, it did have a real prominence in the early 2000s. There was researchers at Oxford talking about it. The reason that AI has become a part of this is a lot of the AI researchers came up in this philosophy and they see AI as a true existential risk. And if you're concerned with the welfare of humanity, and then you start to engage a lot with a technology that you think maybe doesn't have a huge risk, but nonetheless a substantial risk of killing everyone, then that becomes the central concern. I think one thing though that I wanted to just lay out, far be it from me to come out swinging for effective altruists, but like any movement, it's a spectrum. You have people who are real fanatics. And when you hear their viewpoints and the trade-offs that they would make to save a maximum number of people, that kind of utilitarian belief system, it sounds totally nutty. But I talk to a ton of people who identify as effective altruists who are completely normal. I was going to say rational, but that feels really loaded in this context. But they're thinking about it really deeply. They don't sound like religious fanatics. They don't seem like people who are members of a cult. And it's not like all these people are getting together and talking. It's a movement, it's widespread. There are different groups.
  Leah Feiger: And when you apply it to AI, there is something that I find there's an appealing aspect where you're saying, look, we're just saying that we should be concerned about where this technology is going to go and what it could do to people. That's the baseline, right? Is that a pretty decent summary?
  Zoë Schiffer: 100 percent. And if you thought that you were working on a technology that might, 10% chance, kill everyone, the rational thing to do would be to quit and warn everyone about it. So in that sense, what Jacob Coxon did isn't completely shocking. But it has been very interesting to see this go fully mainstream. We have seen AI researchers who are really concerned about AI safety go on Tucker Carlson, Joe Rogan. And again, the Venn diagram of effective altruists and AI safety researchers, it's not a perfect circle, but there is a fair amount of overlap.
  Leah Feiger: So OK, Sam Altman, OpenAI, Dario Amodei, Anthropic. Dario has talked about AI being a risk for the world. So would he be in the EA category?
  Zoë Schiffer: OK, this is a very, very loaded question. So Sam Altman, CEO of OpenAI, worked with Dario Amodei for a while. They were both at OpenAI. Then Dario quit, along with a small group of OpenAI researchers and started a new company specifically to try and build AI safely to benefit all of humanity, etc., etc. And they didn't feel like they could do that at OpenAI. So there you get a clear split between, and this is not how they identify, but this is how people categorize them. Sam Altman kind of represents the growth at all costs, capitalist view of artificial intelligence. Dario represents more of the effective altruism, let's do it safely. We have to be the ones to win because only we understand the real risks involved. Dario has said, I believe, in more recent months and years that he doesn't identify as an effective altruist. That said, a lot of the beliefs do overlap. And I think that you do get a sense from talking to Anthropic researchers that a lot more of them appear to fall into the effective altruism philosophy.
  Leah Feiger: Are there Slack channels for this? Are there parties? I'm so curious, now that this has escaped into a mainstream lexicon, I'm very interested beyond just the Pentagon being like, be pro-American, say no, how this is going to kind of continue.
  Zoë Schiffer: Yeah. I mean, yes, there are group chats. I'm sure there are Slack channels. There's a place in Berkeley called Lighthaven, that's kind of a conference center/group house co-op situation that hosts events. Last time I was there, the possibility of sleeping in one of their bunk beds came up. I was like politely decline forever. No, thank you. But anyway, it's interesting because, as this has gone more mainstream, like you said, the Trump administration has come out and they have really used effective altruism as kind of a overall term to say these people are almost religious fanatics, and they're anti-growth and they're anti-AI progress, and thus they're anti-American. Where of course there is a lot more nuance here. And I think a lot of these people are just trying to take AI safety really seriously. But now you're seeing effective altruism becoming kind of like a slanderous term to say we can't and shouldn't take these people seriously.
  Leah Feiger: Why is this controversial? What you're saying doesn't sound that bad. We're just trying to save humanity?
  Zoë Schiffer: Yeah. Yes. I think one reason is that it's all about trade-offs. And when taken to the extreme, you can see how if your goal is to maximize good or minimize harm, there is a certain amount of harm you would be willing to take, or be a part of, I guess, in order to save a greater number of people. And I think people, somewhat rightly so, get a little freaked out that in an effort to eliminate the possibility that greater harm will be done, effective altruists would be willing to do somewhat harmful things in the present moment because they think they're protecting people from something that could happen later on. And I think the other part, and this is one that I definitely am thinking about, is if you think that you are the only company that's taking AI safety seriously, then you "winning," quote, unquote, the AI race is really important. And what wouldn't you do to win if you think the stakes are humanity level extinction?
  Leah Feiger: I kind of think that everyone involved in this conversation sort of needs to enroll at Oberlin for a philosophy 101 class. That's maybe what I'm feeling right now.
  Zoë Schiffer: I think Oberlin is what got us into this mess, perhaps. Oberlin philosophy specifically.
  Leah Feiger: 102. Philosophy 102. It's the next chapter. Now that you've moved on from Slavoj Žižek, what comes next?
  Zoë Schiffer: Oh my gosh. Moving on, it's been a few weeks since Meta released its new AI assistant Muse, and honestly, it's been kind of an unexpected hit for the company. It is still topping the app charts. But the reason that I wanted to talk to you about this is a really interesting situation played out in the Bay Area a few weeks ago. The designer, a freelance designer and lettering artist, Jessica Hische, who was hired by Meta to help design the Muse logo, told her newsletter subscribers like, "Hey, I did this project." In fact, I think the title of the newsletter was “New Work! (Don't be mad at me).” She kind of knew she was going to get some flack from people. But she basically said, "Look, Meta hired me to do this project. Ultimately, I think a big company should be commended for hiring an independent artist instead of just generating a new logo. Here's how the design process went. Hope you like it." That was kind of the message. One of her newsletter subscribers posted it on Threads. You forgot this was a thing. Threads is still very much a thing in the design world.
  Leah Feiger: I did forget. I fully forgot.
  Zoë Schiffer: And people flipped out. She had people saying they were disgusted with her, disappointed, that they'd followed her since she was in art school and they could never support anything she did again. People were posting photos of her house and speculating about her finances. She felt like the entirety of Threads was like a mob attacking her for three days for having come out and said she was doing this work. Yeah.
  Leah Feiger: That's really intense.
  Zoë Schiffer: Yeah, it was a lot. It was really interesting to talk to her because she actually has kind of complicated feelings about AI. She said her husband has worked in this field, and they actually had some disagreements because he was really excited about the technology. It honestly was very AI wife coded, if you read that article. And she was like, "Well, no, this could be really bad for commercial art." Artists who, to support their less highly paid artwork, work for companies occasionally like she does, this could be detrimental. And in fact, that is what Jessica Hische does. She has two shops in Oakland. They're really cute. One is kind of a stationary store. The other one is a space where she hosts kind of kid art workshops and classes and stuff. And she basically fully funds those by taking on corporate work. But she said in the past couple years, her corporate work had really dried up, and that she wasn't getting as many inquiries as she usually does. So when Meta came to her, she was like, "I didn't even feel that conflicted about it. I was like, Oh my gosh, I want to offer one of my employees a health plan. I need to not be so stressed every time Gusto pings me and says I have to run payroll. Yeah, I'll take this." She was paid. Guess how much you're paid if you redesign the Muse logo.
  Leah Feiger: I mean, it's Meta. You should be paid half a million dollars in equity.
  Zoë Schiffer: Yeah, you would think. No, she was paid $12,500 more or less.
  Leah Feiger: No. No, that's the worst part. That's actually the worst part of this whole story.
  Zoë Schiffer: I know. Yeah. She was like, "I thought it was kind of amazing for two days of work." Because it was a sprint. She was just talking about designing the M at first, and then she got a bit more for doing the full typeface of Muse. But yeah, it wasn't even that much. And again, she's using the money to support her small businesses in Oakland. But she said there was a huge contingent of people on Threads who seemed like they were really AI purists, and fundamentally were like, "There is just simply no excuse for working on this technology." And they really saw her decision to work on it as a signal of support.
  Leah Feiger: I'm mixed here. I don't know. Go after the company, not the artist is sort of how I feel. But I get that that's not as juicy or exciting as yelling at someone personally on a social media app.
  Zoë Schiffer: A social media app that is owned by Meta, to be clear.
  Leah Feiger: Really important point there. Yep. Does she regret it?
  Zoë Schiffer: It's interesting. I asked her that question and she said ultimately no. She kind of weighed like, "Do I regret talking about it?" But she said ultimately she felt like she was doing a good thing for artists by commending a company that had hired an independent artist. She was like, "These big tech companies are not going to hire artists if they feel like that artist is then going to go slam them on social media. Ultimately, these companies exist and there can be some sort of symbiotic relationship with them," which I'm personally sympathetic to.
  Leah Feiger: It's not an Uncanny Valley episode if I'm not at least mentioning or thinking about “Silicon Valley,” the TV show. And I can't not think about the artist in the show who is designing Gavin Belson's box and what that took. And I just remember watching it and being like, why are you putting yourself through this? But also everyone needs a salary. It was painful. Oh, wow. That's a complicated one. That's a little bit tough to hear.
  Zoë Schiffer: Coming up after the break, we're sharing our WIRED/Tired picks for the week. Stay with us.
  [break]
  It's time for our WIRED/Tired segment. Whatever is new and cool is WIRED. Whatever passé thing we're totally over is Tired. Leah, what have you got for us this week?
  Leah Feiger: OK. I want to start by saying everyone should subscribe to Wired.com and everyone should keep reading the news. That's it. No buts there.
  Zoë Schiffer: Great.
  Leah Feiger: But what I would add is it's tiring. This is a very long election cycle. We just finally finished primary season. We are now in midterm season. There is news from Iran and China and Russian plagues every single day. And sometimes it's really OK to stop scrolling and turn the news off. And I do want to remind people of that. Which brings me to my WIRED, which could help you do that. I am not being paid to say this, though I kind of wish I was, I guess. But there's a new season of Great British Bake Off on Netflix, and I'm really, really enjoying it. And it's just comfort and cozy, perfect for October, highly recommend. So read a couple WIRED articles, get nervous about the state of the world, watch an episode, and then go back and do some other fun things. Zoë, what is yours?
  Zoë Schiffer: OK, I love that. Before we get into it, what is your average screen time, do you think? Related. It's related.
  Leah Feiger: Oh. I would say we're in the five-, six-hour range.
  Zoë Schiffer: OK. Last week I was at 14.
  Leah Feiger: Whoa.
  Zoë Schiffer: I was on my phone. I know. Isn't that terrifying? My mental health was horrific, and finally I was like, is this linked? I feel so silly.
  Leah Feiger: Whoa.
  Zoë Schiffer: Yes. Yeah, I know. I know. It's really, really sick.
  Leah Feiger: That's a lot. I'm worried for you.
  Zoë Schiffer: This week I'm down to seven hours. I've started checking it frequently, so more screen time as I check my screen time. OK. My TiredIRED is, I feel like every modern book I have read, this is a gross generalization, but first-person present tense is ubiquitous. This is not used to be how books were written, and now you open your average kind of literary fiction and that is how it's written, really tired of it. George Saunders, great writer, teacher, et cetera, has talked about writing in, I think he calls it third-person ventriloquist. You're in the person's head, but it's still third person. So the narration sounds like the person kind of thinking out loud. And I read a short story that he wrote that kind of felt like this style, I think it's called “Victory Lap,” for The New Yorker last night. He is just a true artiste. I really respect him. I really like everything he says and writes about the creative process and just being a human in the world.
  Leah Feiger: I love that. I'm going to go read. Great recommendation. Are you going to go watch Great British Bake Off?
  Zoë Schiffer: No, I will not, because I still don't own a TV. But I do feel like someday I'm going to get back there and that will be a show that will feel really good. That's our show for today. We'll link to all the stories we spoke about in the show notes. Adriana Tapia produced this episode. It was mixed by Pran Bandi, who's also our New York studio engineer. It was fact-checked by Matt Giles and Daniel Roman. Kate Osborn is our executive producer, and Katie Drummond is WIRED's global editorial director.
  Comments
  Back to top