All-In 的2026年预测
Chamath Palihapitiya × Jason Calacanis × David Sacks × David Friedberg
嘉宾们对宏观经济的核心判断，是一个异常激进的美国增长预期：Sacks押注2026年GDP增长5%，Friedberg押注4.6%，Jason押注4%-5%，Chamath则称5%只是下限，随后又表示6%“并非不现实”。 这套逻辑包括通胀回落、AI驱动生产率提升、边境管控导致劳动力稀缺、退税、加速折旧，以及截至6月可能降息75-100个基点。Chamath称美国经济像一根“盘簧”，但也提醒，政府停工的影响可能让短期GDP数据失真约150个基点，公布值可能低于潜在增速。
加州拟议中的5%财富税，甚至在尚未获得公投资格前，就已经成为人口迁移和资产定价风险。 约850,000个签名和预计800万美元的竞选筹款，可能将该提案送交选民；Jason预计，一旦进入公投程序，就会出现“逃离潮”，随后经历11月投票、诉讼，并可能在2028年推出另一个版本。超级投票权公式尤其严苛：嘉宾以Google为例，Larry和Sergey对一家约4万亿美元公司的52%投票控制权，可能被估值到约1万亿美元，实际税负远高于5%。
铜是Chamath最明确的硬资产押注，因为电气化、数据中心、芯片和武器需求叠加，按当前轨迹到2040年全球供应将短缺70%。 他预计铜价将“绝对呈抛物线式上涨”，并选择一篮子关键金属作为表现最好的资产。反向交易是碳氢化合物：电气化和储能正在制造一座“融化中的冰山”，在他看来，油价达到$45比达到$65更有可能。
AI威胁的是规模达3万亿-4万亿美元的“软件工业复合体”，但未必会先摧毁知识工作的总量。 Chamath称，软件支出中许可费只占5%-10%，维护和迁移约占90%；智能体可能大幅压缩这些高利润池。Jason认为，白领入门岗位的阶梯会消失；Sacks则援引杰文斯悖论，预测更便宜的代码和医学扫描会带来更多需求，并称“失业叙事不仅错了”，最终还会带来就业增长。
资本市场的主线，是IPO大复兴叠加IP授权交易，后者日益替代传统并购。 Sacks预计新上市公司将带来数万亿美元新增公开市场市值；Jason预测SpaceX、Anduril、Stripe、Anthropic和OpenAI中至少有2家提交上市申请，并认为可能出现一笔超过500亿美元的AI收购。Chamath认为，“传统M&A实际上已经死了”，对于敏感技术，未来将出现数千亿美元规模、速度更快的授权与人才交易。
政治版图正向两边的民粹主义倾斜，民主党中间派和科技行业都陷入夹击。 Friedberg预计DSA将巩固其对民主党的接管；Sacks则称，真正有竞争力的众议院席位不到20个，现任议员主要面临来自左翼的挑战。在外交政策上，Chamath认为单边的Trump主义将取代门罗主义时代的假设；Sacks坚持认为委内瑞拉行动与新保守主义战争有根本区别，因为它“没有入侵、没有占领、没有国家建设”，而Jason强调，即便是有限行动，也可能灾难性失败。
Polymarket、Huawei和Amazon，是最具辨识度的公司层面上行押注。 Friedberg认为Huawei和SMIC将跑赢西方预期，并称随着预测合约通过Robinhood、Coinbase以及潜在的Nasdaq扩散，Polymarket将“取代媒体、取代市场”。Jason预计Amazon会成为第一个“企业奇点”，机器人对其利润底线的驱动将超过人类，不过Sacks开玩笑说，Amazon最终可能靠AWS和自由现金流取胜，而不是靠Jason的自动化逻辑。
下行资产篮子覆盖加州豪宅、美元、石油和传统媒体。 Sacks预计，财富税不确定性和交易成本将打击加州房地产；Jason提到联邦债务可能再增加约2万亿美元，以及拟议中的军费预算增加50%，这都将压制美元购买力。Friedberg选择Netflix作为输家，前提是它未能完成对Warner Bros.的收购；否则，输家将是传统媒体整体，因为独立创作者和深厚的竞争性内容库仍在持续令内容和分发商品化。
1. 加州5%财富税威胁已开始重估居住地价值
Chamath估计，明确离开加州的朋友，其净资产合计约5000亿美元，另有25个名字随时可以列出。他仍打算“留下来战斗”，但也承认，许多居民正在对冲风险，因为该提案可能让预算假设中预计征税的财富缩水一半左右，从而削弱它原本要资助的社会项目。
对Jason而言，创始人面临的核心问题是流动性：一名成功创业者可能需要就私人公司股份缴纳5%的税，却没有现金支付，随后公司股价归零，而税务负债依然存在。再加上即使2026年败北，某种版本可能在2028年卷土重来，这种前景“把我推过了那条线”，促使他离开。Sacks另行强调，即便公司随后归零，税款仍然欠缴。
超级投票权条款是最严苛的案例。嘉宾以Google为例，将Larry和Sergey合计约52%的投票权套用到约4万亿美元市值上；相比此前提到的每人1000亿-2000亿美元财富，两人可能各自被认定拥有约1万亿美元身家，结果在融资出售股票并缴税后，征税规模可能相当于实际财富的25%-50%。
程序时间表本身就是2026年的催化剂：约850,000个签名、预计800万美元的筹款行动，4月左右确认是否取得公投资格，11月投票，随后进入诉讼。Friedberg预计提案无法进入选票，Sacks表示同意；Chamath则回应：“在加州，100%。”但这句话在对话中没有明确指向。Polymarket给出的概率为69%，Jason认为如果取得资格，最终通过概率约40%。Jason提出的替代方案，是在政府“修好这个装钱的桶”之后，进行普通资本利得税或所得税改革。
2. 繁荣预期在政治跟上之前，先冲向5% GDP增长
Sacks所谓的“Trump繁荣”，始于公布的通胀率2.7%和核心CPI 2.6%，两者都比预期低40个基点；与此同时，第三季度GDP增长4.3%，贸易逆差降至2009年以来最低。他还指出，裁员数量较11月下降50%，而11月又较10月下降约50%；标普500则不断创出新高。
居民端催化剂按时间表到来：汽油价格下降、按揭成本减少约3000美元、实际工资增加超过1000美元、截至6月可能降息75-100个基点，以及4月的大额退税，受益于更高的标准扣除额和覆盖小费、加班及Social Security的税制变化。
这些预测有意采取大胆口径。Sacks选择2026年GDP增长5%，Friedberg选择4.6%，Jason选择约4%-5%，Chamath则称5%是下限，随后进一步表示6%“并非不现实”。Chamath说，在民主资本主义制度下实现6%的增长，相比历史上与中国式协调增长相关的水平，将是非同寻常的。
他们对政治赢家的选择，暴露出尚未解决的认知缺口：Friedberg选择DSA接管民主党；Chamath为任何打击浪费、欺诈和滥用的人留下机会；Jason看好34岁的“Mamdani时刻”，并将J.D. Vance列为第二选择。Jason认为，Trump在推进海外干预、提出军费预算增加50%的同时，放任工人阶级的不满继续暴露。
3. 民粹主义夹击中间派，也把科技变成靶心
Sacks选择民主党中间派作为今年的政治输家。Cook Political Report和Sabato’s Crystal Ball都认为，真正有竞争力的众议院席位不到20个；因此，多数民主党现任议员担心的不是共和党，而是更年轻、类似AOC的初选挑战者。为了守住左翼侧翼，即便名义上的温和派也会继续向社会主义靠拢。
Chamath对2025年主流叙事的反驳基于事实：Mamdani只赢得一场市长选举，但Virginia和New Jersey选出了更多中间派人物；随着进步派话术增多，民主党的全国支持度反而恶化。他称，认为进步主义广泛成功的说法是“事实，而不是感觉”，并表示这种模式在全国范围内的回报“在分类上、在数学上”都不存在。
Friedberg预计，科技将成为最大的政治输家，因为AI和科技财富如今同时吸引来自两边的民粹主义愤怒。左翼认为科技与Trump站在一起；右翼则仍记得审查、影子禁令、封禁和取消银行服务。Jason进一步引用3名共和党资深参议员的报告：在多年感受到科技公司及其领导层不尊重之后，他们已将其中数家公司和领导人视为不可信。
Sacks仍认为MAGA是科技的天然盟友，因为它捍卫财产权和创新——“想活命就跟我来”。但他也表示，和解需要会面、承认问题并道歉。Jason接触的那些参议员正是这么要求的。加州财富税倡议可能通过明确硅谷面临的进步派替代方案，加速这种政治重新结盟。
4. Trump主义打破新保守主义剧本——前提是行动保持有限
Chamath将门罗主义列为2026年的政治输家，因为更广泛的Trump主义已经取代了它。他的框架包括半球主导地位、针对贩毒集团的选择性干预、更严格的移民控制、确保战略资产安全，以及交易型双边关系；“单边主义”和国内经济韧性，正在取代多边主义假设。
Jason认为，Trump对委内瑞拉的行动以及外交政策威胁都属于新保守主义行为，并提到有关Colombia和Greenland的言论，以及政府更广泛的干预主义。他承认上述行动执行得异常出色，但认为，那些竞选时警告选民提防干预主义政客的人，一旦在任总统采取类似行动，听起来就会不一样。
Sacks通过3点区别拒绝这一标签：新保守主义行动需要入侵、占领和国家建设。他将抓捕Maduro描述为一场持续3小时、没有美国人员死亡的行动，随后与既有权力结构合作，而不是扶植María Corina Machado。“诺贝尔奖不能让人掌权，手里有枪的人才能让人掌权。”避免全面去复兴党化，也就避免了类似伊拉克的叛乱。
Jason的反驳值得保留：成功并不会消除尾部风险；如果有12名Delta Force成员被俘，或50名美国人死亡，讨论会完全不同。Sacks的回答是，担心的结果并未发生，并将这套战略概括为“出手，移动”。双方都讨论了谈判作为下一步的可能性；Jason另行预测，Trump访问China可能基本解决这场对峙，包括Taiwan问题，而且不必让任何一方输掉。
5. 铜、Polymarket和机器人领跑商业赢家榜
Friedberg将赢家押注分给Huawei和Polymarket。Huawei与SMIC的深度合作将跑赢西方预期；Polymarket则已从一个古怪的博彩场所，演变成实时新闻和洞察来源，预测市场正通过Robinhood、Coinbase以及潜在的Nasdaq扩散。他压缩后的判断是：“取代媒体，取代市场。”
Chamath选择铜，因为在数据中心、半导体和武器系统中，铜仍是最便宜、最实用、最具延展性的导电材料——“到处都是，到处都是，到处都是”。在一个多边安全秩序弱化的世界，各国需要确保本国供应，但按当前路径和速度，到2040年全球供应将短缺约70%。
Sacks选择IPO本身。过去多年，随着上市公司转为私有，上市公司总量不断收缩；他预计这一趋势将决定性逆转，多笔成功发行将带来“数万亿美元新增市值”。Jason相关的判断是“超级IPO之年”，公开市场投资者终于能够买到那些已经在二级市场被大量交易的资产。
Jason选择Amazon作为第一个“企业奇点”：人类员工数量大致保持不变，而机器人越来越多地驱动利润底线；Zoox和密集的履约网络，则能让Austin的订单在数小时内送达。Sacks预测，Jason的股票判断最终会因为无关的原因而正确——更高的自由现金流或AWS增长——让他成为“运气爆棚的家伙”。
6. 智能体进攻软件维护和白领晋升阶梯
Friedberg预计，当浪费调查与巨额、无资金覆盖的养老金义务相撞时，州政府将难以融资。真正危险的信号，是官员选择维持支出，而不是纠正滥用。他支持固定缴款制，而不是固定待遇制：建立类似401(k)或Australia制度那样可见、有资金支撑的账户，而不是不断累积、背后没有资产的负债。Jason补充说，强制个人缴款约10%-14%，可能让美国人拥有更多主动权。
Chamath所说的“软件工业复合体”，对应着一个规模约3万亿-4万亿美元的年度经济体。初始许可费只占5%-10%，维护和迁移约占90%。如今，智能体可以完成那些“战术性、琐碎、不太性感的工作”，从而激进压缩现有SaaS收入，同时为拆解3亿美元级遗留部署的新公司创造高利润机会。
Jason将美国年轻白领列为商业输家，因为企业发现，自动化入门级任务比培训Gen Z更容易。Google、Uber和Coinbase已经在用更少的人完成更多工作，企业晋升阶梯最底部的2-3级正在消失。他的建议很直接：掌握AI的年轻人仍然能找到工作；缺乏动力、又拒绝使用AI的人将陷入困境。
Sacks通过一位朋友提供了反例：这位朋友参加过一场约50名CEO的聚会，所有人都表示仍在招聘初级工程师，只是数量减少；他们把原因归结为后COVID时代人才质量、性格和执行功能变弱，而不只是AI。Jason承认这是多因素解释，包括文化、父母财富和疫情冲击，但仍坚持认为，自动化正在吸收的，恰恰是历史上分配给应届毕业生的底部三分之一任务。
7. IP授权成为M&A的逃生通道
Sacks认为，编码助手和计算机工具使用正在跨过新的质量门槛，但也承认其中一些兴奋可能只是炒作。能够本地访问文件并采取行动，让这一刻有点像“聊天机器人在2022年底进入2023年时的状态”：突破已经显现，经济后果则会在随后一年持续累积。
Friedberg认为最大的一笔“交易”将发生在地缘政治领域：他预测Russia和Ukraine将在2026年达成和解，经济和政治激励正在汇聚。这将带来更高的地区稳定性，并成为全球大国重新定位自身的更大重置的一部分——即使Trump是在第二年而非第一天促成这件事。
Chamath预计，将出现数千亿美元规模的IP授权型M&A替代方案，参考案例包括Google–Character AI、Microsoft、Nvidia收购Groq，以及嘉宾提到的Scale AI交易。Facebook试图以25亿美元收购Manus，正好说明了限制所在：中国正在审查的不仅是技术转移，还有关键研究人员的转移。授权可以立即转移人才和IP，以不完美的税务处理换取惊人的速度。
Jason仍预测，一家Mag Seven公司将与xAI、Mistral、Perplexity或Anthropic完成一笔超过500亿美元的收购。Chamath同意，现金充裕的公司将面临交易压力，但直接收购可能需要至少3年的跨国反垄断审查；因此，最终规模达到1000亿美元级别的交易，更可能被设计成经过优化的IP授权。
8. 逆向地图指向Iran、AI劳动力和SpaceX
Friedberg的推演始于一场由生活成本问题驱动的起义：推翻Iran的阿亚图拉，建立独立的民主国家。意外之处在于，Iran可能曾经在稳定地区秩序；它退出后，UAE、Saudi Arabia、Qatar、Yemen各派和Somaliland之间的竞争可能加剧，尤其围绕影响力、巴勒斯坦问题的责任和资源展开。他预计Middle East会变得“更加糟糕”，而且未必以Israel或Iran为中心。
Sacks预测，AI会通过杰文斯悖论增加、而不是减少对知识工作者的需求。更便宜的代码会释放企业委托软件开发的总量；AI辅助放射学会让扫描更便宜、更常规，但医生仍需负责提示、解读和验证。随着应用场景扩张，效率会降低单位劳动需求，却会增加工作的总量。
Chamath预测，SpaceX不会进行传统IPO，而是反向并入Tesla，让Elon Musk在同一张资本表上整合这两项标志性资产；Neuralink和The Boring Company理论上也可以纳入更广泛的控股结构，但具体押注是SpaceX与Tesla反向合并。
他的第二个货币领域逆向判断是，央行将为资产负债表寻找一种超越黄金和Bitcoin的新型加密资产或新范式：可互换、可交易、私密且由主权国家控制。隐私可以让各国的持有量对盟友和对手保持不透明；新的加密方案则能对冲量子芯片在未来5-10年内攻破当前安全体系的风险。
9. 生产率和税收政策扩大资产繁荣
Friedberg再次选择Polymarket作为表现最好的资产；Chamath选择一篮子关键金属；Sacks选择不断扩张的科技超级周期。Jason选择博彩综合体——Robinhood、Polymarket、PrizePicks以及潜在的Coinbase——逻辑是，降息、更强的盈利和家庭闲置现金将推动投机活动增加。
节目播出时公布的宏观数据进一步强化了这些判断：美国生产率据报飙升4.9%，创近6年来最高；Atlanta Fed对2025年第四季度GDP的估计，则在1月5日至1月8日期间从2.7%跳升至5.4%。
Chamath立即补充了限定条件：政府停工可能让公布的第四季度数据减少约150个基点，使潜在GDP增速更接近4%。此外，移民减少已将非农就业数据的基准从100,000-150,000个岗位重置为40,000-50,000个。收入分布较低四分位的工资正在加速增长；据报Ford有5,000个技师岗位，最高薪资可达160,000美元，这些都支持“盘簧”判断。
Sacks强调，符合条件的资本设备可以享受首年100%加速折旧。飞机、拖拉机、发电机和工业机械的需求已经非常强劲，Caterpillar和Siemens等供应商因此受益，同时商业投资和GDP也得到提振。如果按揭利率下降100-150个基点，工资增强，加上对企业购买住房的限制，也可能让住房复苏更多转向个人买家。
10. 房地产、石油、美元和媒体构成下行篮子
Sacks选择加州豪宅，因为财富税不确定性正在叠加严苛的交易成本。讨论提到，San Francisco超过2500万美元的房产需要缴纳5%的豪宅税，Los Angeles也有5%的豪宅税；此外还有一项6%的经纪费，买卖双方可以协商。他希望看到的反向交易，是公投倡议失败后出现一轮“死猫反弹”，足以缓解悬压并推动房产成交。
Chamath预计碳氢化合物表现不佳。无论气候政治如何发展，电气化和储能都在持续减少石油的可服务用途；这不是突然崩塌，而是一座“融化中的冰山”。按每桶价格计算，他认为$45比$65更有可能。
Jason选择美元的各种可投资形式。他预计年内债务将再增加约2万亿美元，而拟议中的军费增加50%可能进一步施压。即便美国经济强劲增长，美元价值仍可能受到挑战，黄金、白银和铜的上涨趋势正体现这一点。
Friedberg选择Netflix，前提是它未能完成对Warner Bros.的收购。竞争对手拥有深厚内容库，创作者越来越不喜欢Netflix“成本加10%”的经济模式，内容也在商品化。如果Warner Bros.完成交易并补充内容库，他的输家将转为传统媒体整体，因为独立创作者正利用YouTube和直接分发渠道挑战娱乐和新闻行业的既有公司。
11. Nuclear可能错过窗口，也可能满足太阳能无法承载的需求
Chamath仍然“做空核能”，理由是经济而非科学。复杂的审批流程会将大型电站推迟到2032-2035年；届时，他预计太阳能、储能、煤炭和石油会把边际电力成本推向零。大型反应堆届时将失去经济意义，而小型模块化反应堆可能在市场转向后才到来：“数学上无法自洽。”
Friedberg的反驳，是需求曲线发生了位移。China正在迈向约8太瓦的发电能力，而United States接近1太瓦；如果要追上，可能还需要增加2-3太瓦。他质疑太阳能的土地需求、建设速度和耐久性是否能独自承担全部负荷，并指出China一些大型太阳能项目已经开始拆除。
因此，分歧在于时点和规模。Chamath认为，核能会错过当前已经可见的市场；Friedberg则认为，未来大幅增长的电力需求将迫使各国采用混合发电结构，最终形成核能能够服务的拐点。
12. 公民调查者成为可变现的媒体类别
Sacks最期待的治理趋势是“去中心化DOGE”：让各级政府的公共审计常态化，并“让一千个Nick Shirleys绽放”。他对比称，Pentagon至少还在努力通过审计，而California官员据称在阻挠对无家可归者支出的审查。举报人和独立调查者应当让公共支出变得可理解，而不是把监督完全留给机构。
Friedberg和Chamath对媒体作出同样判断。公民新闻正在从被动记录事件，转向主动带着摄像机进入现场、揭开事件真相。Substack、GoFundMe以及YouTube和X上的收入分成，提供了一种可再投资的商业模式：越重要的调查能带来越多观看和资金，进而支持更深入的后续工作。
Jason的“第一修正案审计员”视频，是其中最具娱乐性的样本：创作者在公共场所拍摄，引发投诉，并测试警方是否会捍卫宪法权利。对于传统娱乐，Sacks和Jason都在期待Christopher Nolan的《The Odyssey》；Friedberg称Homer的原著很糟，Chamath对电影前景押注为零，而Jason则在Polymarket上询问，能否做空它的票房。
This is what we need. Let him go. All right, here we go. This is Jason in the corner warming up. Two. Shut the up, Friedberg. It's my show. [laughter] Three. Two. All right, everybody. Welcome back to the number one podcast in the world. The podcast I, Jason Calacanis, named, created, and I'm the executive producer for life. With me, my three bestie friends, Chamath Palihapitiya, our dictator; David Friedberg, our Sultan of Science; and David Sacks, who's now made his way down to Austin. Welcome, brother. Let's go shooting. Let's get those beef ribs.
David Sacks, how are you settling in to the great state of Texas? Everybody wants to know. Let's go shooting and get those beef ribs.
I'm loving the 70-degree weather. Is it like this all year round?
This is a wonderful time of year. You miss the 2 weeks when it goes to freezing temperatures. We have 10 days of freezing temperatures, and then we have 80 days of 100-degree temperatures, but you'll be on a yacht in Italy or somewhere during that, like the rest of us. That's basically all you need to know about Austin: get out during the summers because it's brutal. Everything else is fantastic. But in all seriousness, you're here. You've moved. You've domiciled in Texas.
It happened in December. We closed on a new house, moved in, went to the DMV, and I signed a lease for an Austin office for Craft. It's done.
Nice. I'll get you a dentist and whatever else you need.
I got a doctor, too.
Okay. Does this mean I have to bring Moose back? This has been the big discussion in our house. Does this mean we lose Moose, or do we have you in custody?
He needs significant acreage to run around. We know that.
Oh, he does. He misses you, and we'll definitely bring him by for a visit.
We'll be playing backgammon and smoking cigars.
You have to come by. I cannot wait. Chamath, what about you guys? Are you guys going to come down?
Matt and I started the process in December. We are coming to check things out. We have not made any final decisions, though.
Okay. Shout-out to our boy Ro Khanna for driving everybody out of the state. Here's the funniest thing: we're all in the chat group discussing the California wealth tax, whether people are going to leave or not. Chamath is making a big show: “I'm going to stay and fight. I'm not leaving my home. They can't drive me out.” Meanwhile, I get a call from my broker, who says she's helping Chamath find a place.
Uh-oh. What's going on?
Is Chamath doing a backdoor trade? I'm shocked that Chamath tells you one thing and is doing another.
He's at least hedging his bets.
He's hedging his bets. A lot of people are hedging their bets. Sergey is in Florida. I'm sorry—Larry's got a beautiful place in Florida. I saw Governor Abbott reached out to you.
That's right. He welcomed me to Texas on X. A lot of people did.
Michael Dell did.
Ted Cruz did, too. It was a real welcome reception. It's funny: I never got anything like that when I was in California.
The politicians were never embracing you.
They weren't embracing you.
Yeah, they were embracing me. I don't know why.
Who knows? I put this on X. When you look at our friends who have all explicitly left, it's about half a trillion dollars of net worth, which I think is very bad for the long-term budget of California. I can pick about 25 people just off the top of my head. If you think about all the people who will stay and fight, I still intend to stay and fight, but if we're forced to look down the barrel of an asset tax with God knows what methodology, it's the only thing that's united everybody on the left and the right. Even Reid Hoffman thinks this is insane.
I think a lot of other people will leave. It's probably half the total wealth that the budget estimated would be available to be taxed will be gone.
Wow. I mean, these are—I don't understand.
That has huge implications for the social programs and the general budget of California.
Listen, I think this is going to be a topic throughout the year because it's not going away. They're gathering signatures right now.
I agree.
We're going to find out in April whether it's on the ballot. It's possible they don't gather the signatures, but they only need about 850,000 of them. If it gets on the ballot, we'll know in just a few months. There's going to be a huge freakout, and I think there will be a lot of people who say, “I can't take the risk.” They're going to leave the state.
I think there's going to be a rush for the exits. There will be all sorts of repercussions from that. Then there will be an election in November. Obviously, we'll find out whether it passes or not, and then there will be legal challenges. This is going to be a saga. I don't think this is over by a long shot.
Quite frankly, even if it's beaten in 2026, I think a lot of people expect that some version of this comes back in 2028. That's the thing that pushed me over the edge in terms of leaving. I don't think this problem is going away. It's very difficult if you're an entrepreneur with a good idea to start building here, because if you get stuck in success with a bunch of illiquid stock and have no way to pay 5% of that value, you're going to bankrupt your own company. It just doesn't make any sense.
What if your company goes to zero the next year, which can totally happen with private companies? You still owe the tax bill.
Yeah, I don't know how that reverses itself. It's a terrible idea. I was talking to Ro Khanna and Eric Swalwell, who we had on the program, and I told these guys, “You dopey Democrats have to stop fraud first before you start seizing people's assets or start a discussion about raising taxes.” We had Nick Shirley on last week, who was doing his investigative journalism. Shout-out to Shirley.
How do we sell to the American public that we want to seize their assets while sending it out the back door to fund fraud? That makes no sense. It's a leaky bucket, so fix the bucket first, and then let's have an honest discussion about what the right tax rate is. Why do this with a unique tax? Why not just add a point to the capital-gains tax or income tax and have that discussion?
That's a great point. You could also do what Bill Ackman suggested, which is stop allowing these margin loans. There are a lot of people who live off margin. Is it a good idea? Yes and no, depending on the asset base you have. If the tax laws changed, we would all change our approach.
You mentioned that Larry and Sergey have left the state, and they're probably getting dragged for that. But one of the reasons why I think they kind of have to is because of the super-voting stock provision in this thing. The way they calculate the value of your stock is not based on its liquid market value. If you own super-voting shares, they multiply your ownership of those shares by the market cap of the company and deem your shares to be worth that.
For example, Larry and Sergey, I think, combined have voting power of about 52% of Google. What's Google worth these days?
Four trillion—4 and a half trillion.
Okay. I think they're very wealthy guys. I think they're each worth whatever—$100 billion or so.
I think $200 billion combined.
But now their net worth will be deemed to be roughly $1 trillion each, not, let's call it, $200 billion. The 5% tax for them is more like a 25% tax on all their net worth. It becomes 50%, because you'd have to sell more than that to overcome the drag of selling. To generate $25 billion or $50 billion of net worth, you have to sell twice that, because you have to pay taxes on that, right?
What is the point of having that super-voting provision in there? It's totally punitive and vicious.
All right, let's just go with a quick prediction here. I'm going to call an audible. We're going to get into the prediction show, folks. We've been delaying this prediction show because the world is moving at an incredible pace, and the news is, let's just call it what it is, intense. We have a lot of stories you all want us to cover. We will cover them, but we're going to start with our prediction show.
Just a lightning-round prediction: does this seizure tax—I'm going to call it what it is, a seizure tax—go into effect or not? Does it pass or not? Friedberg, you're first. Yes or no? Give us a percentage, Polymarket.
Does it get on the ballot first, and then does it pass?
We know it's going to get on the ballot. I think we agree with that, but not necessarily.
Oh, we don't know. So then I'll make it a 2-parter. Does it get on the ballot? Does it pass?
Fine. Chamath, you seem ready to go. Go.
No. Friedberg, go ahead.
I don't think it's going to get on the ballot.
Oh.
I agree with that.
In California, 100%.
All right, here's your Polymarket, everybody. Shout-out to Shane.
No, this is to make the ballot. This is to make the ballot.
This is the question.
Will the billionaires’ wealth tax make it onto the California ballot? It’s currently lightly traded at 69%.
But you see how much lower it was? That spike happened after Ro Khanna elevated the issue. When it became a cause célèbre among progressives and Bernie Sanders weighed in, it spiked from—what was it, like—45% to 80%.
Okay. So when Ro Khanna committed political seppuku, it drove it up. Maybe he had a bet. Maybe he placed a bet.
I don’t think it’s political suicide for him. There are only 2 ways it doesn’t get on the ballot, right? One is if the SEIU, which is the union that supported this proposal, doesn’t have the money to pay to collect the signatures.
But you would think that they would, right?
Mhm. $8 million or so is roughly what it costs to gather the signatures for these types of things. Clearly, they can find 850,000 people in California who support it if they’re willing to put the effort in.
The other possibility is that the powers that be—let’s call it Gavin Newsom and the machine—are able to negotiate with this union to get them to stand down. I don’t have any insight into that because that’s obviously Democratic politics, not Republican politics. But Friedberg and Chamath might have some information. You have some insight. I don’t know. We’re going to leave it at that.
But if it does get on the ballot, what do you think the odds are that it passes?
I think it’s going to be a really important
40%
moment for people to vote for the ability to be industrious and have agency. Or what did Mamdani say? It’s rugged individualism versus
Collectivism, also known as communism, as in “collect.” We collect your assets, Chamath.
He said that we’re going to replace the fragility of rugged individualism with the warmth of collectivism.
I mean, at least he’s saying it out loud. I like these new politicians just telling you straight up what we’re doing. Well, his new housing commissioner—have you seen this?
Yeah. The white lady’s like, “White people must suffer.” Then she had these things where she’s like, “I see white babies in the airport and it makes me mad.” I’m like, really? This is crazy.
What man broke up with this lady? Because that dude has created a monster.
Didn’t she say something like, “We need to introduce a new relationship between white people and property and their property rights,” or something?
Yeah, exactly.
I mean, they’re saying it out loud now. Then I guess she had a mental breakdown and started crying when they came to see her.
No, no. What happened is they pointed out that her parents own a multimillion-dollar home, and then she broke down in tears.
Oh, her mom’s a white supremacist. I didn’t realize her mom was a white supremacist. That explains it all, because this is all about her mom.
I feel bad for her parents.
Oh, God. Can you imagine one of your kids going and doing this and saying, “We have to go collect your houses”?
I mean, listen, let’s do it. Let’s do what everybody wants to hear: our predictions for 2026. We’ve been pushing it off, and we’re going to give it to you right now.
We always like to start, since we got into politics here on the program at some point, with the biggest political winner. Last year’s prediction for the biggest political winner for 2025, to remind everybody: Friedberg said young candidates. Well done. Gavin, who was on the show sitting in for Sacks, who was busy joining the administration, said Trump and centrism would be the biggest political winner. Chamath, you said fiscal conservatives and those who asked for restrained spending. It was a good thought, and I said Gen X and elder millennials—the J.D. Vance, Tulsi, Sacks group.
Let’s go around the horn here. Friedberg, who do you think will be the biggest political winner of 2026? Friedberg, your chance.
Democratic Socialists of America, the DSA. Just like the MAGA movement took over the Republican Party, I think the DSA is taking over the Democratic Party. I think that’s the move we’ll see solidified in 2026.
Okay. Tight is right. Well done. Chamath, who do you have for biggest political winner in 2026?
Whoever is going to fight waste, fraud, and abuse at the federal, state, and local level.
Got it. So it’s an open lane to anybody.
It’s an open lane. It’s a political gambit that I think will work really well in ’26.
Very nicely done. David Sacks, I can’t imagine who you would pick as a political winner in 2026.
Well, I’m going to say that the Trump boom is going to be the biggest political winner of 2026. The good economic news started breaking out before 2025 was even over. We have 2.7% inflation and core CPI at 2.6%; both of those are 40 basis points below expectations. We had 4.3% GDP growth in Q3 and the lowest trade deficit since 2009. The Challenger, Gray & Christmas report out today showed that job cuts dropped 50% from November, which was itself down about 50% from October.
Give us a number for the boom. What is it going to be?
Hold on. The S&P 500 keeps making record highs. People are paying less for gas. Mortgage costs have fallen by $3,000. Real wages are up over $1,000. By June, I predict we will see more rate cuts, possibly 75 to 100 basis points. Big tax refunds are coming in April, thanks to a bigger standard deduction and no tax on tips, overtime, and Social Security.
I think there is so much good economic news coming, and it’s already started. I think it’s going to have a huge impact not just on the economy, but also on political perceptions for next year.
Pick your GDP for this boom year: 3%, 4%, 5%, or 6%, Sacks. I’m pinning you down.
Pick 4%.
If you make it a prediction, then I’d like to pick as well. Okay, I’m going to let Sacks go first.
I’m going to go for 5%.
Within a rounding error, I believe 5%. Sure, sure. Plus or minus? Chamath, you want to pick a number? You said you did.
I think the lower bound is 5%. I think the upper bound is 6.2%.
Wow. Incredible. Well, just to put that in perspective, if we print 6%, the only country in the modern world that we think of as a quasi-peer competitor that has printed 6% is China, in a period where it had complete and total coordination and domination of a federal, state, and local economy. The fact that we can do it under democracy and capitalism is outrageous. Friedberg, do you want to take a stab at that?
2026 growth: 4.6%.
Okay. I was going to go with between 4% and 5% as well.
In terms of my prediction for the big political winner, I went back and forth between the emperor’s apprentice, Darth Vance—J.D. Vance—or the Mamdani moment. I said Darth Vance for a couple of reasons. He is out there defending Trump, and he’s surging on Polymarket and in the polls. He’s obviously the co-pilot of MAGA. He’s done a great job of being in the second seat. He is not usurping President Trump, which would be a big political mistake. He’s really navigating being the co-pilot there.
He is the most popular politician, clearly, at Turning Point USA, and he is the OG in the America First, America Only moment.
But I’m going to give the edge to the Mamdani moment. He’s 34 years old. We’ve got Ro Khanna at 49, pivoting into a socialist, and I think that’s because Democrats believe the easiest way to win in 2026 is to go full socialist. Trump has, I think, given this lane because he’s forgotten about the working man and woman in America.
Net disapproval for Trump on the economy: 58%. Inflation, despite what Sacks is saying there, is still closer to 3% than 2%. Trump just announced he wants to increase the military budget by 50%, while people are still complaining about their healthcare. Trump has turned into a complete neocon, bombing 7 countries this year and threatening to take over Colombia and Greenland. Who knows if that’s Trump being Trump or if that’s reality, but Trump becoming a neocon was not on anyone’s bingo card.
I think Trump may have incited and given a bunch of fuel to the Mamdani moment by not addressing the American people’s needs and going for international interventionism. Okay, now—
That was like 3 different answers right there.
My answer is clearly Mamdani. My second place—though I like to always explain my thinking—is Darth Vance.
J.D. Vance, biggest political loser. Last year’s predictions: I said Putin. Gavin said Putin. Chamath, you said progressivism, and Friedberg, you said the pro-war neocons would be the biggest political losers.
Let’s get into who we think will be the loser this year. Sacks, why don’t you start? Who’s your big political loser in 2026?
Well, I said Democratic centrism, or Democratic centrists, which is sort of the flip side of you guys saying that socialism or progressives are winners. There are 2 reasons for this.
One is because the socialist ideology has ascended among the Democratic base, especially the young people who support Mamdani and things like that. Unfortunately, our universities are woke madrassas that have done a terrible job educating these students and have brainwashed a lot of them into this woke ideology.
But also, there are so few House districts anymore that are truly competitive. Both the Cook Political Report and Larry Sabato’s Crystal Ball say that there are fewer than 2 dozen House races that are genuinely competitive going into 2026. That’s because of gerrymandering and so on.
So, if you’re a Democratic incumbent who is in one of these districts—like all but a couple dozen of them—your only real threat to losing your office is from the left, right? It’s some young AOC type coming up to challenge you.
And so you don't want to expose your left flank. Even the Democratic moderates have been shifting further and further to the left, and so you see this AOC–Mamdani effect happening there. I'm kind of in the same camp as you guys: this is not a good trend. I put this as the biggest political loser.
Biggest political loser?
Democratic centrism.
Got it. Chamath, what do you got?
Can I just go back and nitpick with you a little bit? Why do you think progressivism didn't fail? The only reason I ask you that is, outside of a few pockets of progressivism—specifically Mamdani—if you look at the elections in Virginia or the elections in New Jersey, those are more centrist than progressive.
If you look at the general approval rates of Democrats, they trended consistently down through 2025 as Democrats congratulated themselves about leaving centrism and embracing progressivism. The more talking points around progressivism that emerged, the poorer they performed. I know your perception, or some people's perceptions, may be different based on one localized win, but if you look at the broad trend, it didn't work. I just put that out there as the facts and not the vibes.
Well, I don't want it to be true, just to be clear. I would—
I don't want it to be true either, but it does seem—
No, but you—like, it was wrong and it wasn't wrong.
Well, it might be a jump ball. That might be the best way to describe it, because you have effect and—okay, just look at the numerical numbers. The trends were horrible. Wherever the Democrats started, the more progressive talking points they added, the poorer and poorer they performed. The approval ratings went down, and the disapproval ratings went up.
I'm not saying that they didn't win a mayoral race. They did do that. I'm just saying, broadly speaking, nationwide, has the Democratic Party's embrace of progressivism, at least at the federal level, paid off over 2025? I would say categorically, mathematically, not.
Okay, now going to 2026, what is my biggest political loser? What I would say is the biggest loser of 2026 is the Monroe Doctrine. I think that when historians look back on the Trump presidency, they're going to rewrite it. People have tried to minimize Trump's worldview as a Trump Corollary. I don't think that's what this is. They even try to minimize it by calling it the Donroe Doctrine. I don't think that's what this is.
I think that there is a clear Trump doctrine that trumped the Monroe Doctrine, and I think that is the political loser because there is a huge body politic that has been built around the Monroe Doctrine. How do we view wars? How do we view our spheres of influence? How do we view economic multilateralism versus unilateralism? All of that is out the window.
We view this as hemispheric dominance. That's Trump. We view it as proactive and, in very specific cases, interventionist. We intervene against drug cartels. We control immigration. We secure vital assets. That was not really the scope of the Monroe Doctrine. We have more transactional relationships, quite honestly, which allows us to react in the moment. So I think the Monroe Doctrine is the biggest loser of 2026.
Friedberg, your biggest loser of 2026?
My biggest political loser of 2026 is the tech industry. I think AI and tech wealth have become the lightning rod for populism on both sides of the aisle.
I think the right is fracturing a bit, where this alliance between tech and MAGA seems to be getting a really strong challenge from the more populist movement. In the same sense, the left is turning hard on tech because of tech's alignment with the right. I think we're going to see, in the midterms, a really big referendum against the tech industry coming out of a populist win.
Good. Populist win.
Can I tell you guys a little story from yesterday? I had a meeting with 3 very senior sitting senators before I flew back to California. These were Republican senators, and, Friedberg, I was surprised: exactly what you said.
There are a couple of companies that have exacerbated their frustration. They view those companies—these tech companies and the tech leaders of these companies—as just not trustworthy, and they've largely been ignoring these guys for a long time. They're pretty frustrated with it. So, to your point, it is palpable.
Look, I can tell you that the natural ally for tech is with MAGA because we still believe in property rights and innovation. If the Democratic Party is truly going progressive, which means socialist, they want to rewrite your relationship to property rights, whatever that means, and impose wealth taxes, realized-gains taxes, and all the rest of it. I don't think tech has that much of a choice.
As Arnold Schwarzenegger said in one of those movies, “Come with me if you want to live.”
“Come with me if you want to live.” Yes.
Get in the chopper. But let me say this: the reason why there's anger on the populist right is because they remember the censorship and the deplatforming and the shadow-banning and all that kind of stuff. I think there just needs to be some meetings, some truth and reconciliation, between some of these tech leaders and some of these conservative influencers.
Guess what, David Sacks? I know 1 guy who can help make that happen. I would like to host some of these meetings in 2026 and get these people together because I think the tech companies have either realized their mistake or, in a lot of cases, were pushed into it by the Biden administration.
Yeah, they had a gun to their head.
They had a gun to their head. Now, I also think that one other mistake they've made is, quite frankly, they've been donors to only left-wing causes. If you listen to Mike Cernovich's account, he's like, “Look, guys, you banned us. You cost us our livelihood for years.”
Debanked us.
Debanked us. Where's the restitution? Or at least start giving some support to our conservative culture.
By the way, you nailed it on the head. The senators that I talked to, that's exactly what they want. They just wanted an apology. Just be honest and say you did it.
This asset-seizure tax proposal in California, and the conversation about other states, I think, is bringing a lot of people to that table. At least on the tech side, contrary to what people may think, it may actually be doing a lot more positive for the right side than the left by putting this forward. This may actually be catalyzing a big change in Silicon Valley.
Yeah. If you think about it, if you were Zuckerberg or you were one of the Google executives and the FBI was telling you, “Hey, we need you to take care of these censorship issues. We need you to label these things,” et cetera, it puts you in a pretty tough situation if the FBI is calling you, if you're trying to do M&A. Now, M&A under Trump is on fire. We'll talk about that more in our prediction show.
For me, I was going back and forth on my biggest political loser between these dopey Democrats who are centrists and the new neocon Trump. I don't know if Trump will continue these neocon ways, so I'm going to align with Sacks here that the centrist Democrats are going to be this year's biggest political losers.
Now that you've mentioned twice that Trump is a neocon, I have to respond to this.
No, you don't. You don't have to, but go ahead.
Okay, look, the problem with neocon regime-change operations was, I'd say, 3-fold. Number 1: the invasion. You have this giant invasion, land armies, huge numbers of people getting killed. It took months or a year. It took them like a year to get Saddam, right?
Number 2: you then have an occupation because whoever you put in power only stays in power if you have American GIs there pointing the guns. So you end up with a 10- or 20-year occupation.
And then, third, you have nation-building, which means you end up spending trillions of dollars basically trying to turn them into us so that our troops can leave. That was the mistake of Afghanistan and Iraq.
But look, what has Trump done that is like any of those things? There's been no invasion, no occupation, and no nation-building. This war, I guess, with Venezuela, if you want to call it that, was a flawless operation. It lasted 3 hours.
I mean, I woke up and it was like the meme where it's, “Wake up, honey.” “Wake up, honey. Trump's won another war.”
It was over before it even started. They went in there and basically captured Maduro. No Americans were killed. It was an absolutely flawless operation, and they are bringing him to justice.
By the way, he begged for it. He's on tape talking a lot of smack at rallies, saying, “Come get me,” calling the Americans chicken, and so on. In any event, he was begging for it. We can go into a lot more of the reasons for doing it and defending it, but I just don't think this is a neocon policy.
In fact, it's the Democrats who've been calling to put in this Nobel Prize winner that you interviewed, Friedberg, right? What's her name? María Corina Machado.
Machado.
Machado. Okay, anyway—
Héctor Elizondo. I think he did somebody from the Brat Pack.
No, she was in The Running Man, I think. Anyway—
Exactly.
But—
You're talking about the Cuban-Venezuelan actress.
I thought her name was María Conchita Alonso. Okay, María Corina Machado. Sorry. That's just what I thought of: Miss World Venezuela 1975 winner, María...
Look, the Democrats are criticizing the administration for not putting her in power. But here’s the problem: Nobel Prizes don’t keep people in power. Men with guns keep people in power, and she doesn’t have the men with guns. So it would be American GIs. There would have to be guns to keep her in power, and the administration has not done that. They’re looking to basically work with the existing regime.
The big reason why we got sucked into Iraq is that whole de-Ba’athification process. We didn’t just get rid of Saddam; we took out the entire elite of the country, which created a huge insurgency. So there’s been nothing like that. That is what I’m trying to tell you in this case: it’s a whole different paradigm. We’re going to need a new name for it. Maybe Chamath is right. It’s definitely not neocon.
Not neocon.
Yeah, we might need a new branding for it. It depends on whether Trump is cosplaying a neocon when he says he’s going to take Greenland, when he says he’s going to take Colombia next, and when he says he’s going to take Cuba. He’s certainly playing the character of a neocon publicly here. Who knows? That could be Trump positioning himself and anchoring future negotiations.
But if I was telling you before the election, when you were saying, “Hey, do not let certain people become president—Nikki Haley, et cetera—because they’re neocons and they’re going to go to war with Venezuela. Are they going to go to war with Iran?” Well, that’s exactly what Trump has done. And as flawlessly as our troops did—and, my lord, we have the greatest military ever—just an incredible job. Shout-out to them. To do this and lose no American lives was unbelievable, and it says something about the dedication of these individuals. But things can go wrong no matter how good you are. We could be sitting here right now with 12 captured Delta Force members. We could have 50 dead Americans.
Where’s the war? But if that did happen, this is where you have to be intellectually honest. If we were dealing with a situation where they didn’t pick up the target and we had lost American troops—and, God forbid, they had taken hostages—we would be sitting here with a much different discussion, so we have to be very careful. I give Trump credit, but that is an equal possibility, or certainly a nonzero possibility. Things can go sideways.
You sound like when Sam Harris was saying, “Imagine if COVID actually killed a lot of people. Then the conversation would be different.” Well, it didn’t. There was no war here.
And let’s hope there isn’t. This is why I give Trump a lot of credit. He has been strategic. I do give you that.
Stick and move. Stick and move is the game.
I pray that he can continue.
And by the way, first of all, even if we took Greenland, it’s not going to be a big deal. There are 30,000 people who live there. But I think it’s more likely that we’ll make a deal. I think we’re going to make an offer they can’t refuse.
Okay, Ken Howery, shout-out to our guy Ken Howery. Maybe you can make a deal. Let’s make them an offer they can’t refuse.
Yeah, I mean, how much could it cost? There’s nothing there.
Biggest business winner for 2025: it looks like Friedberg picked robots and autonomous hardware—the year of the robot. I think if you include robotaxis there, you nailed it. Certainly, next year will be the year of Optimus, 2027.
Chamath, you said dollar-denominated stablecoins. I think, given what we’ve seen with regulation, that was spot-on. Gavin, shout-out to our friend Gavin. He said big businesses that use AI thoughtfully. Another great one.
I picked Tesla, which is at an all-time high, and Google, which, of all the Mag 7, was the biggest winner. They did 65%. I know both of those—we did. I think all of us crushed that one. Who do you have, Friedberg, as your biggest business winner prediction for 2026? Go ahead, Friedberg.
I couldn’t decide. My number 1 is Huawei, which I’ve mentioned in the past, out of China. I think Huawei’s effort to partner with SMIC to go deeper in the chip stack—they’re just firing on all cylinders. I do think, keep an eye on Huawei over the next year. It’s going to outperform expectations, at least Western expectations.
The second is Polymarket. I think Polymarket has evolved from being this one-off, quirky prediction market to actually providing insights into current events and the news in a way that none of us anticipated.
And I do expect that, after the deal we saw with the NYSE, all of the exchanges—and we’re already seeing this with Robinhood and Coinbase—will follow suit. We should expect something from Nasdaq this year. Dina Friedman talked to us about this, but I do think that prediction markets could become not just markets, but also news. I think Polymarket is in such a position to have a breakout year.
Okay, great one. Chamath, biggest business winner for 2026, after yourself. Who do you got?
Yeah, it’s hard. I mean, I will pick me.
Yep, we already did that one.
That one’s coming.
It’s already in the books. Don’t hurt your elbow. Don’t hyperextend your elbow. It’s already in the books.
Nick, we need to use that meme of Obama giving himself a medal.
I will pick copper.
Okay, copper.
We are still completely underestimating how short we are in terms of the global demand-supply dynamics of a handful of critical elements that we need. Again, in the Trump doctrine view of the world, that is no longer as multilateral as it was. We need unilateral national security. If you look through that lens, the asset that is set up to go absolutely parabolic is copper.
The reason is that, at least as it stands today, it is the most useful, cheap, malleable, conductive material that we have. That material manifests in everything from our data centers to our chips to our weapon systems. It’s just everywhere, everywhere, everywhere.
Right now, Jason, we are on a path by 2040 where we will be short about 70% of the global supply at the current course and speed. I will pick copper.
Sacks, what do you got? Biggest business winner of 2026. Your prediction.
I said the IPO. I think 2026 is going to be a big year for IPOs. I’m not going to say which ones. I think there are going to be a bunch of them, a bunch of successful ones. I think we could see trillions of dollars of new market cap created by public companies.
Totally.
For a while, people were concerned that the number of public companies was shrinking. Public companies were actually being taken private. This is going to be a big reversal of that trend, so I think this will be part of the Trump boom.
I love that one. Great one.
I went with Amazon. Since I nailed the Mag 7 for last year and picked the highest performer with Google—and I placed a bet on that—I’m feeling pretty good about it. My prediction for 2026 is that Amazon is going to have a massive year as they continue to replace humans with robots.
Here’s a chart for you. I’ve been talking about this a little bit. I think this is the most important company to watch because Zoox’s self-driving is working. They’re making great progress with it. When you look at this chart, you can see they’re essentially flat in terms of hiring humans, and they’re surging in deploying robots. They did their whole PR and communications strategy of calling them cobots and donating to Toys for Tots, et cetera.
I think they’ll be the first corporate singularity—which is to say, the first company to have more robots driving its bottom line than humans. So that’s my prediction for 2026: Amazon. I’m going to place a bet on that.
My prediction is that Jason is such a luck box that he’ll end up being right about Amazon, but not having anything to do with the reason he gave.
Yeah, right. He’ll hire a billion humans to go do something, and it’ll be for a different reason. I am a luck box. That is true.
It’ll just be better free cash flow, and all of a sudden more people will be using AWS, and Jason will be right.
Well, I actually think the reason is, if you just think about it—I don’t know if you guys have had this experience—but with the delivery business, which was kind of a dog for a long time, obviously AWS is crushing it. But that delivery business here in Austin—we get everything the same day. You’re going to experience this, David, because of the geography here and the ability to have depot centers very close. Everything you order on Amazon comes within the same day.
You know, I’ve noticed that already. I’ve already started ordering things from Amazon. You’re right; it was all same-day. I’m like, “Wait, what is going on?”
That’s what it is. We have a lot of space 15 miles outside of the city center, with 20 centers, and they built these huge warehouses. So you just get everything within 4 hours. They’re like, “We’ll be right there.” It’s very bizarre.
So, Jason, you’ll buy, like, virtues at 8:00 a.m., and it arrives by noon.
Yes, absolutely. And when you put your order in for ethics and morality, it just never shows up. I don’t know. It’s lost again. Who knows? It’s just incredible. We’re just both ends of the spectrum here.
It’s like, “Hello, customer support. I ordered a moral compass 2 years ago. It still hasn’t arrived.”
Biggest business loser.
I love xAI, man.
It hid behind the $20 billion.
It did.
Absolutely.
All I know is I got my beak wet.
So, I hope your compass never shows up.
You got your beak wet on that deal.
Well, Sundeep came to Besties, and he wanted to collect all the Besties. We all got to place a little bet, and then—
Oh, by the way, I divested my Groq shares as part of joining the government back in February or March.
Oh, God. So, what would that cost you? What was the last triple-up?
No, it was a small position that came from Groq acquiring Sundeep's old company. We didn't invest in Groq, so it wasn't big, but the point was—
Yeah, it's just another example of you sacrificing for the country, which I give you a lot of credit for. It's good for Sacks to put that on the record.
Yeah. Well, it's just so ridiculous because I'm accused of somehow doing this job for money when it just keeps costing me money.
It's negative money.
xAI just raised an up round at twice the valuation from the last round, and we had to divest that too. No, no, we didn't sit it out. Hold on. Let's be clear: We didn't sit it out. We divested it.
So, you place the bet, and then before you get to collect the ticket, you lose the last double-up or triple-up. The economic cost to Sacks will probably exceed $1 billion by the time he leaves.
Personally, this is crazy. People need to know this. It’s a very important thing to put on the record. I think everybody should know. All these dumb reporters don’t get it, but Sacks has sacrificed financially an enormous amount to work on behalf of the government and the people. That’s really amazing.
It's fine. I wouldn't even say anything about it if it weren't for the fact that we got these mainstream media reporters lying and saying the opposite—that somehow this job is making me money.
Right?
I wouldn't say a word about it otherwise.
Business loser 2025. Gavin said federal government service providers, I guess because of DOGE and them being held to the fire to give us a better deal. Old-guard defense contractors like Boeing and Lockheed was yours, Friedberg, and Chamath, you said the Mag 7 would see a decrease in record concentration. I said OpenAI, which would see a peak valuation, and obviously I got that wrong. What do you got this year, Friedberg? Who do you think is going to be the biggest business loser of this year?
To follow up on Chamath and my conversation at the Christmas dinner, I think these state governments are going to have a real problem finding financing. What's going on with the exposés underway on waste, fraud, and abuse in state agencies is going to lead to a conversation that causes folks to question the long-term solvency of their operations. The response won't be, "Hey, let's cut out the waste, fraud, and abuse." The response is going to be, "We've got to keep it going," and that is what is going to give people fear.
If they responded equivocally to the discoveries that the governments made, then I think there's an opportunity to continue to borrow and access capital markets. But I do worry a lot about state governments borrowing. I think the other thing that's going to hit the fan this year in state governments is all of these unrealized pension liabilities.
I think when these numbers start to come out this year, and a lot of people are now digging deep into them, folks are going to wake up and be like, "Holy, there's a ginormous hole in these states and their obligations."
And why does the government have to do these pensions? Why can't we do superannuation like Australia does? I love that pick.
That's called a defined contribution instead of a defined benefit. If you get this defined-contribution model and then just have good management, it all works out. So, this is basically the problem with Social Security. Social Security is a defined benefit, and then all the money just doesn't sit anywhere. It doesn't exist. If it was a defined contribution, it's just a liability.
It's just a liability.
And let me give a shout-out. If it was a defined contribution like Invest America Accounts, or Trump Accounts as they're being called, and you put the money in and see how much you have, and every year you track it, that's what your retirement is going to be, just like we might have with our 401(k)s or our IRAs, that is a system that actually has true solvency. Otherwise, it becomes this runaway train of liability. That's effectively what the states have set up, and it's very dangerous.
And if we were to take government out of it and then every American just had to put 10%, 12%, or 14% into their retirement account, and it was forced, you'd have happy people who feel some agency in their lives, which the people of Australia do. Chamath, who's your loser for 2026?
I will pick the software industrial complex. These are the companies that sell licensed SaaS to the corporations of America. It's about a $3 trillion to $4 trillion-a-year economy, and that is separated into 3 buckets.
Bucket 1, which is the smallest, is the initial licensing. That's probably 5% to 10%. Buckets 2 and 3, where all the money is made, are what's called maintenance and migration: How do I just maintain this big, bulky license that I just bought for $300 million, as an example, or how do I migrate it from product A to product B? Those last 2 buckets represent 90% of all the dollars in revenue that's generated in software.
Because of the advancement of these models and the advancement of these technological techniques that we are all uncovering—building agents, building systems—I think you're going to see that total economic opportunity shrink and contract aggressively. The companies will still be able to do their business; it'll just be at a much, much lower incremental revenue. The customers will be able to do their business, they'll have a lot more flexibility, and a lot of upstarts will have opportunity.
I'm speaking my book, obviously, but I'm seeing it on the ground. When you have a company building software, 80% to 90% of the business has basically migrated to disrupting maintenance and migration patterns. I cannot describe how much opportunity there is. It's very tactical, mundane, not very sexy work, but it's incredibly lucrative.
And so I expect that thing is going to shrink. It's going to impact SaaS companies, particularly public SaaS companies, quite severely in 2026.
What do you got, Sacks, for your biggest business loser of 2026?
For me, we already talked about this, but it was California because of the wealth tax and also the onerous regulations driving business and capital out of the state. I hope you guys are right that it does not make the ballot. If it does, I think there will be a panic and rush for the exits.
Regardless, there are 2 major refineries closing by the spring. Higher gas prices will be the result. I just think that the politicians are not doing a good enough job in California dispelling the fears and the actual hostility of the business environment.
That's a great one. I went with young white-collar workers in America. I think they're going to be the biggest business loser. I think it's getting really hard for them to get entry-level jobs. I'm seeing that all over the place because companies are having an easier time just automating with AI than training up Gen Z graduates. That's my belief. That's why I launched Founder University on 3 continents.
If you're a young person, you've got to be resilient. You're going to have to be self-reliant, independent of what Mamdani says about collectivism. It's easier to use AI than it is to train people up, and we don't have professional development. That needs to come back. I just did an interview with the CEO of McKinsey.
The big challenge in corporate America is that they're taking out the bottom 2 or 3 rungs and automating stuff. We really need to develop young people so that they have a path to take the CEO jobs eventually, and I don't think they're going to have an easy time doing that. That's why I think all young people should start companies. I am talking my own book. I am talking about founder.university. Please apply in Japan, Saudi, and America. We're going to help you build companies. Thank you for my promo.
Jason, I got a text last night from a friend in response to your comment about young people not being able to find jobs because of AI, which is a statement you've made a couple of times. He went to a roundtable of 50 CEOs of public and private companies and asked everyone if they're hiring junior engineers. Everyone said they are still hiring them, but not as much as before, because during COVID every college lowered the bar on admissions and the talent just isn't as good anymore.
My friend went on and said, "You should talk about this to counter Jason's point about AI taking young people's jobs. We see this with financial analysts and salespeople we hire. The Gen Z kids are all really challenging to hire because of cultural issues, not because we're not hiring them due to AI. So, we try to hire older people primarily to fill those roles."
There's a really interesting point that he was making. I texted a couple of other friends to ask their opinion, and I've heard this concurrence, which is that a lot of people think recent grads out of college—and this may be a COVID-era phenomenon—just don't seem to have the temperament, the motivation, the organizational skills—
Executive function, yeah.
Executive function. And, by the way, some of our friends I've talked to who have kids graduating from college, there's even a conversation that none of these kids are motivated to get jobs or to make money. There's a very weird phenomenon in the youth right now.
This may be a COVID phenomenon, and it may be a cultural thing that's part of the long arc of what's going on in our society. Or it may be a socialist trend, or it may be a populist trend, or it may just be that people have gotten too wealthy and the nation has truly split and you can't climb the ladder anymore.
There’s a bunch of things going on here, but I do not think—and I think a lot of people are echoing this, JCal—that the challenges young people are having finding employment are purely rooted in an AI and automation phenomenon. But it may be a cultural phenomenon. So I just put that on the plate for you to consider.
I do think it’s both. I do think it is partially what you’re saying: maybe these young folks have either become entitled, or their parents have enough money for them to skate and go sideways and maybe not be as career motivated. It could be a social thing, it could be a COVID thing, it could be all of those, and it’s certainly multifactorial. I just know what I see on the ground, which is so many companies coming to me saying, “We can replace the bottom third of these tasks,” and those bottom third of tasks are typically done by young people out of school. So I think both things are probably true to a certain extent, and time will tell.
I do think it’s going to be challenging and continue to be challenging, and you see that in the numbers from Google, Uber, Coinbase—all these companies are doing more with less. Maybe that’s just the nature of AI. Maybe the first thing you do is cut costs, and maybe the second thing you do is hire people who know how to use these tools. If you are a young person who uses AI tools, you’re going to find a job. If you’re a young person who isn’t motivated and doesn’t use AI tools, you’re going to have a hard time.
Let’s go with the biggest deal. Last year’s prediction for biggest deal for 2025: traditional auto OEM consolidation. That was you, Chamath. I think we haven’t seen exactly the consolidation, but we have seen those businesses come apart, so I think I give you 2/3 of a credit there. Sacks, you said a tidal wave of M&A. I think you’re not wrong there; that has started to happen, certainly. Friedberg, you said massive compute buildout deals. Of course, you nailed that one.
I said consolidation amongst the on-demand economy. That hasn’t happened, but we do see a lot of deals occurring. Gavin said a tidal wave of M&A. I also said Apple would buy Warner Bros. I got that, I guess—that wound up going to Netflix—so I give myself 1/2 point for that. What do you have as your biggest deal for 2026, David Sacks?
Well, I don’t want to get too specific here in terms of names of companies and that sort of thing. But what I would say is that I think there was a breakthrough in the last couple of months in terms of these coding assistants. They’ve been around for a while, but there seems to have been another level of quality achieved just in the last month or so.
You’re really starting to hear—maybe a lot of it is hype—but I think a lot of people are getting very excited about the potential here. Part of it is coding; part of it is just tool use. You can download the programs, so they have access to your file drive and can take actions on your computer. This trend feels to me like chatbots did at the end of 2022 going into 2023, where people were really hyped about it, but then it continued to play out in the next year. So I think this whole coding-assistance and tool-use trend will get bigger and bigger this year.
Friedberg, what do you got for biggest deal?
Russia-Ukraine, I think it’s going to settle this year. I think there are a lot of motivating factors to get it settled this year—economic and other political factors—but I do think it’s going to settle this year, and it’s going to bring a bit more stability to that region. There’s a whole reset underway this year, I think, in terms of geopolitics and where the powers all sit.
Trump can stop that war; he would be 2 for 2. I think this one’s going to settle.
That would be great. He may not have done it on day 1, but if he gets it done in year 2, that’s good enough for me.
Chamath, biggest deal for 2026. What do you got?
It’s not a specific deal, but it’s an approach. I think that M&A cannot happen, and so it’s the IP-license M&A workaround. I think you’re going to see hundreds of billions of dollars of these kinds of deals. This is the deal that Google did with Character AI. It’s the same thing that they did.
Microsoft did it, yeah.
Microsoft did it. It’s obviously what Nvidia did with Groq. Why are these deals happening? If you just look at what Facebook tried to do, they tried to buy Manus for $2.5 billion. Manus was a Chinese company that then left China and essentially rebuilt itself as a Singaporean business.
The Chinese have now said, “We’re going to look at this.” They are going to actively impose export controls. They’re going to actively look at which technologies and even which researchers are working on things that are critical enough that they just can’t go abroad in this existential fight that they believe they’re in with the United States. The United States is in an equivalent position.
All of this leads me to believe that traditional M&A is effectively dead. I think it’s going to be impossible to get a large transaction done. So how will you do it? You’ll do what Sundar did. You’ll do what Satya did. You’ll do what Jensen did. You’ll do what we did with Scale AI. These huge licensing deals basically replace M&A.
I think that, as a deal type, will get better and more refined and tighter and better executed. We were the third or fourth of these kinds of deals, and even the third or fourth iteration was quite good. By the middle part of next year, after you’ve done 15 or 20 of these things, I think the lawyers who work on these things and the accountants will just be bulletproof. The tax treatment is not ideal, but the speed at which you can do them is phenomenal, because the next day somebody like Sundar can be working for Jensen, which is what Jensen wants. They want the talent. Zuckerberg wants the talent working there the next day.
And the IP.
And the IP, of course.
For 2026, I think we’re going to see some massive M&A. It doesn’t matter to me how it occurs, but I do think we’re going to see a $50 billion-plus deal. I think it could be one of the Mag 7—Apple, Meta, Microsoft, or Amazon—going out and trying to buy xAI, Mistral, Perplexity, or Anthropic. One of those 4 comes to mind.
I know most of them probably want to go public and go it alone, but I think an offer could come in during that race among the 6 or 7 large language models. Man, it is a battle where they are moving up and down the rankings and beating each other out. I think one of them is going to go for it. I think it could wind up being Apple.
I mean Meta or Amazon buying Anthropic or Perplexity. I started where you were, but this is why I went to this deal type as the biggest business winner: if any of those companies tried to buy, let’s just say, Anthropic, I think it’s 3 years of antitrust minimum. It’s worse than when Microsoft tried to buy Activision, because that was a niche product, and even that took almost 2 years and 3 or 4 months, if I’m getting it right—or about 2 years.
It’s a huge slog because it’s about global coordination of multiple regulators. You have to get through the EU, and eventually one of them takes the lead position. But in the Microsoft case, it wasn’t just one. You had to navigate China. You had to navigate Europe.
Yeah.
So I agree with you, because I think there are companies with so much cash on their balance sheets that they’re effectively getting debased every day. The markets will start to punish these companies. It just seems like you’re right: there’s going to be a $100 billion transaction. I just suspect it’ll end up as an IP license.
Yeah. And I think President Trump, one of his great strengths is that he moves quickly. Man, what a first year. Whether you like the decisions or not, he makes decisions. I think this is one of the things that Democrats are learning: you have to actually get things done for the American people.
I think he might instruct our government to let M&A be great again, and that would be great for American exceptionalism. These companies do need to merge and continue to grow, and we should go try to get from a Mag 7 to a Mag 17. We need more, bigger companies with bigger footprints taking on global markets.
Most contrarian belief. People like this one. I said OpenAI loses its lead in the AI race, and in fact that has happened. If you look at the arenas and you look at their market share, they are being challenged. Chamath, you said the banking crisis in one of the major mainline banks. Gavin said 1 year of 5%-plus GDP growth at one point over the next couple of years. Well done, Sacks. Friedberg, you said socialism roars back. Another amazing prediction.
What’s your prediction for this year, since you crushed it last year? Friedberg, go ahead.
My prediction is based on the premise that I think there is going to be this revolution in Iran and the Ayatollahs are going to be out. That’s not the contrarian belief. I think that’s the standard belief, and I think that is going to happen, and that’s the premise.
But a lot of people think that Iran is part of the destabilizing force in the Middle East. I do think that, already in anticipation of the turnover with the ruling parties in Iran, there is this brewing conflict amongst the other Arab states. So I think that between the UAE, Saudi Arabia, Qatar, and this faction in Yemen—and I don’t know if you guys have followed, but there’s this kind of emerging independence movement for Somaliland, which is north of Somalia—there may be more conflict brewing in the Middle East than anyone anticipates for this year that will not necessarily involve Israel and/or Iran.
It will actually be amongst the other Gulf states as they vie for influence and power. The contrarian point may, in fact, be that Iran has been a stabilizing force in that region, and that by removing Iran and changing it into this kind of independent democratic state, the conflict will be heightened. There is going to be a battle over who is going to take care of the Palestinians as the two-state solution emerges, and what role Jordan is going to have to play versus Egypt and Saudi Arabia. It is going to lead to a lot of questions about resource allocation. So I think that this year could end up being a little bit nastier than folks anticipate in the Middle East as Iran turns over.
Yes. I said that AI will increase demand for knowledge workers, not decrease it. I would refer you to Erin Levy's post called “Jevons Paradox for Knowledge Workers.” The point of Jevons' paradox is that as the cost of a resource goes down, the aggregate demand for it actually increases because you discover more and more use cases.
I think this will certainly happen with code. In the past, it has been very expensive to generate code. You have to hire engineers, there aren't enough of them, and it's an expensive resource. So the amount of software generated in the economy was limited by that. I think it's going to increase massively now because the cost of generating code is coming down so much.
But there are other examples, too. Take a field like radiology, which is frequently cited as a profession that AI is going to put out of business. That's not what the data shows. The data shows that the number of radiologists is increasing. Why? Because the number of scans that people want to make is increasing. It's true that AI can do some of the work, but you still need a doctor to prompt the AI, interpret the AI, and validate it.
So you get more efficient, the cost of scans goes down, and instead of it being a super-specialty that happens very rarely—that you need a referral on top of a referral to get—it becomes something that's normalized. Everyone starts doing it, and you start getting more and more scans. That leads to better and better outcomes. I think there are going to be a lot of those examples throughout the economy, and we're going to look back and see that the job-loss narrative was not only wrong, but that we actually got job gains.
Okay. For those of you who want to understand Jevons' paradox a little more, you can look at something like electricity, steel, or concrete. When we lowered the cost of those things, people didn't use less of them. They built skyscrapers, and we had more routes for more airplanes to take you on more vacations. That went from being something only rich people did to something everybody did. Jevons' paradox is definitely at work.
Chamath, what's your contrarian belief?
I have 2. I'll give you both, and you can decide. My contrarian belief number 1 is that I don't think SpaceX will IPO. I think it will reverse-merge into Tesla, and I think Elon will use it as a moment to consolidate control and power of his 2 seminal assets into 1 cap table.
Oh my God, I love that. Wow.
Well, he's talked about that before.
He's talked about having a holding company for years to hold the entire collection. You could put Neuralink in there, too, right, Chamath? You could put in the Boring Company as well.
Sure. I'm just giving you my contrarian take. There will be no IPO for SpaceX. I think it will specifically be a reverse merger.
There you go.
The second contrarian take is that I think the central banks will realize that there are limitations to gold and limitations to Bitcoin and will, as a result, seek out a completely new cryptographic paradigm that they can control on their balance sheet, that is fungible, that is tradable, and that is completely secure and private.
I think the reason why that privacy needs to exist is that, for the sovereignty of a country, you need to be in a position where you have assets that are not easily disclosed to anybody else, friends or enemies alike. Separately, cryptographically, if you're going to own a currency, you need to hedge against the eventual risk, in the next 5 to 10 years, that there's a quantum chip that can challenge the existing cryptographic schemes that are used.
For my contrarian belief, I was thinking about going with OpenAI losing its lead and not being the number 1 company again. I do think that will be the trend: that it will continue to give up market share to other players, including Google Gemini, xAI, et cetera. But I'm going to go with a pretty wild card here.
I think the standoff with China is going to be largely resolved, and I think that when President Trump makes his visit there—I don't know if you're going to go on that one, Sacks—I think he should obviously be included in it because of the AI race. I do think that the standoff and the issues around Taiwan are going to be resolved. I think this could be the signature issue of Trump's second term: that we work out a working relationship where both China and America win without one of us losing.
Best-performing asset. Last year's prediction: I said the Magnificent 7, which was up 22% versus the S&P 500, which was up 17%. Chamath, you were long CDS for a potential run on a major bank. You said it was a long shot. Sacks said high-bandwidth memory makers like Micron, which was up 230%. That was a great call. Friedberg, you said Chinese tech stocks and ETFs. Alibaba is up 85%, and the Chinese tech ETF is up 47% versus the S&P 500 at 17%. So I guess that means Sacks, Friedberg, and then myself.
What do you have for this year's best-performing asset? Friedberg?
Polymarket. Polymarket's on a tear: network effects, replacing media, replacing markets.
Congrats to our friend Shayne Coplan and Polymarket. Do you have a best-performing asset for 2026? Go ahead, Chamath.
I would pick a basket of critical metals.
Okay, a basket of critical metals. What do you have, Sacks?
It's getting a little redundant for me, but I just said the expanding supercycle in tech. Again, this is just another facet of the boom. But actually, let me just show you some data that literally just came out. I feel like this is breaking news.
Oh, breaking news.
Nick, can you pull this up? U.S. productivity just surged 4.9%, the strongest reading in nearly 6 years. And the news item is that the Atlanta Fed's forecast for Q4 GDP just climbed to 5.4%. Can that be right?
Yes. Can we get a fact-check on that? 5% to 6% is, I think, where we're going to see it, guys. We're going to see some 6% prints.
The January 8, 2026, Atlanta Fed GDPNow model estimate for real GDP growth in the fourth quarter of 2025 is 5.4%, a significant jump from the previous estimate of 2.7% on January 5.
You have to remember that there's going to be a 150-basis-point correction in Q4 GDP because of the government furlough. So let's say that GDP was probably, call it, 4%. You're going to see a print of 2.5%. You need to readjust that because now all the government workers are back and they're recounted in GDP.
If you look through 2026, there's a handful of things that I think people do not understand. Number 1, all of nonfarm payrolls has been completely reset and rebased, and the reason why is immigration. What used to be a 100-to-150 number print is now a 40-to-50 number print. Why is that important? Because when you look through earnings and you look at the lower 25% quartiles of earnings growth, they're off the charts.
There are all these anecdotal examples now of earnings just exceeding expectations. There was an article on the front page of The Wall Street Journal yesterday about Ford trying to pay mechanics $160,000 a year and having 5,000 openings. So, to Sacks' point, we are a coiled spring. Closing the border, plus adding productivity lifts through AI and other things, has created a growth dynamic in the United States that will really start to show itself in 2026. I think you should not be short the U.S. economy here. It is ready to rip.
Adding to all that, and just giving my pick, I think if we are in a rate-cut environment, the tailwinds keep happening, and people have a little bit of cash laying around, my pick for best-performing asset will be the Robinhood, Polymarket, and PrizePicks gambling and wagering space, because people will be able to have a little cash around to make some bets. You can put Coinbase in there, too, I guess.
By the way, the corollary to what we just talked about is that if you see 5.5% and 6.5% prints and these employment numbers, a lot of this affordability stuff may not be as accurate as we think it is. Everybody right now is trying to figure out where the pockets of unaffordability are. There are clearly some, but those are narrow and they can be fixed. On a broad-based basis, what Sacks says is right: you have this combination of earnings growth, productivity growth, and now this overlay where you have these tax cuts that are going to hit in 2026.
My gosh, asset prices in general, I think, will do well. Now, you'll also potentially see home prices correct because if the president is successful in making sure Blackstone can't buy houses, but on the other side people are earning more and can enter with interest rates that are now 100 to 150 basis points lower, you'll see a boom in housing where it's not corporations buying the houses but individuals. There's a lot of variables here that can break in America's favor. This is why I think 6% is not unrealistic, which would be absolutely nuts.
Okay, let's move on to the worst-performing asset.
Last year's predictions for worst-performing asset of 2025: I said legacy car companies and real estate. Both of those turned out to be correct. Chamath, you said enterprise SaaS and the software-industrial complex. Again, that looked correct as well. Gavin said enterprise SaaS, and Friedberg, you said vertical SaaS. So we had 3 SaaS predictions, and I had legacy car companies and real estate.
Let me give you credit. Let me just give you some numbers to show you: ServiceNow was down 30%, Workday was down 18%, DocuSign was down 23%, Dropbox was down 9%, and Box was down 6%, while the S&P was up 17%. I think it's worth highlighting that it was a challenging year for enterprise SaaS.
Especially with per-seat pricing as opposed to consumption-based pricing. If you have a static number of employees or fewer employees, as many of these companies do, there are simply fewer seats to sell. Their whole growth model was based on land and expand: you land the client, and then you expand the client because they're adding staff. If you're not adding staff, you don't have more people using Salesforce.
Let's talk about our worst-performing asset predictions for 2026. Who do you have, Sacks?
I just said California luxury real estate because of the overhang of the wealth tax and all the things we're talking about.
Yeah, that one hits close to home, I think.
Now, yes, it does. Actually, what I'm hoping for is a dead-cat bounce. If you guys are right that the ballot initiative fails, then the overhang will be lifted and maybe I can clear some real estate.
Clear some real estate?
It's not easy being right sometimes.
What is your discount price on that asset right now? I might make a bid if you give me a good clearance price. I've got $100 million laying around.
I'm not paying $100 million.
Okay, here we go. Bestie negotiations are occurring.
All right.
Give him the bestie price—the buy-it-now bestie price. What's the buy-it-now price?
No, no, no. By the way, San Francisco has that insane luxury tax, which they have in L.A., too. You have to pay 5% on any piece of real estate over $25 million in San Francisco. Do you know that?
Which makes it even harder for these high-end places to trade. So wait, you're saying if I bought Sacks's house for $100 million, I'd have to spend—
Oh, he has to pay 6%.
6%?
You have to pay your broker 6%. They're basically freezing the market.
Okay, not the buyer.
Well, you can negotiate who pays.
The 5% so-called mansion tax has just killed L.A. real estate. You talk to brokers down there, because people used to flip houses a lot more. Now you just can't afford to do that.
Unintended consequences, folks.
Wait, you're telling me taxes slow down transaction volume and reduce the growth of the economy? That's crazy. I had no idea, Sacks. You should write that down. We should do a whole thing on that.
Yeah, maybe we should write it down. I had no idea.
Somebody make a note.
Yeah.
Chamath, what do you think? Worst-performing asset?
I won't say the worst-performing asset, but I think a very poorly performing asset will be hydrocarbons. I just think the trend in oil is inexorable, and it's down. The reason it's down is that, irrespective of your thoughts on climate change, the trends toward electrification and energy storage are just unstoppable.
What that does is shrink the surface area of where oil is useful. It's not a cataclysmic thing, but it's sort of a melting iceberg. Does it see $65 or $45? On a per-barrel basis, I think it's more likely to see $45 than $65.
I went with the U.S. dollar in different permutations of how you can buy it, because our debt continues to grow unabated. I just think it's going to be hard for the U.S. dollar. We've been adding, I think, $2 trillion in debt this year, and if we're increasing—and again, I know President Trump says a lot of things—the military budget by 50%, that means that's going straight to our debt line. It's going to be harder and harder for the dollar.
That doesn't mean America isn't going to do great, but the value of the American dollar is going to be challenged, which we see in people moving to gold and silver, and perhaps copper.
If you could give us your worst-performing asset of 2026, David, what would it be?
It would be Netflix if they don't close the Warner Bros. deal. I do think Netflix's service is being challenged from all sides by deep content libraries, and I think we're seeing a great commoditization happening.
I've also heard directly from folks in Hollywood—the creators of new content—that people would prefer not to work with Netflix. They only pay creators cost plus 10% now. As a creator, you're actually better off not doing deals with Netflix anymore. Their content library is going to shrink because of the natural economic forces underway.
Alternatively, if they do close on Warner Bros., I think they've got some good runway in terms of that content library, and they'll be fine. In that case, my worst-performing asset would be traditional media stocks. I do think they're going to underperform. There's just such an incredible variety of high-quality content emerging from independent creators who are leveraging their own distribution platforms through YouTube and others. Traditional media is going to continue to be deeply challenged, as we've seen, for example, in the news segment with the rise of citizen journalism.
That's what I would say.
The Netflix observation, I think, was well founded, because in terms of expanding their library, they just did a deal with Bill Simmons, I see, and with Barstool Sports to move over their sports shows, podcasts, and video shows from YouTube. They're going to take them off YouTube and put them on Netflix exclusively. That's a really interesting trend to keep an eye on as well.
Before we move off the asset discussion, just on best-performing assets, there's 1 category we didn't talk about that I think is interesting: assets that qualify for accelerated depreciation—capital equipment.
Yes.
That includes things like planes and so forth. There's now 100% accelerated depreciation for certain kinds of capital equipment because of the One Big Beautiful Bill. It kind of goes along with some of the tax cuts that we talked about, but that is making those markets super hot right now.
Absolutely. Try buying a plane these days. It's very difficult.
No, Jason. It's not planes. It's like Caterpillar.
No, no, it's tractors. Yeah, it's generators. It's Siemens. This is why Siemens stock is through the roof. All of this capital equipment—you get to write it off 100% in year 1. It's creating a massive infrastructure build-out in the U.S.
Absolutely.
There are some companies that are huge beneficiaries of this, obviously, because they're the sellers of the capital equipment. Of course, this is 1 of the reasons GDP is going up: people are actually investing in business again.
The corollary of the statement that if you reduce taxes, the economy grows.
Most anticipated trend of 2025: I said the wrath of Lina Khan ending, and M&A and IPOs being back. I get some credit there. Chamath, you said the end of the deep state. I think you get a lot of credit there. Gavin said, “AI makes more progress per quarter in 2025 than it did in 2023.” That's a great 1. Friedberg, you said the nuclear power build-out. I think you get some credit there, too.
I'm not sure. We have some work to do there. I'm massively short nuclear.
You're still short nuclear, but I know the guy, Howard Lutnick. I talked to him.
I talked to Howard Lutnick as well.
He said he's all in on nuclear.
Friedberg, can I tell you why I'm short nuclear? I think we're in the very delicate part of the cycle where they've missed the window. By the time they deliver working SMRs at scale, the problem is that the marginal cost of electricity will effectively be zero. It will have gone to zero because of a combination of solar and storage, as well as coal and oil.
It's just in a very delicate place where large-form-factor nuclear reactors make zero economic sense by 2032 or 2035, which is when, through the Byzantine permitting and building process, they get them done. The SMRs, by the time they get them done, may not be able to meet the market either. I think it's a very complicated moment for nuclear. Not scientifically—economically, it just does not hang together mathematically.
I think that's if you assume no shift in the demand curve. If you look at China going to 8 terawatts of production by 2040, or whatever it is, while we're sitting at 1 and not moving, we are going to have a big catch-up to do. The question is: can we really build out 2 to 3 terawatts of electricity generation? How are we going to build out 2 to 3 terawatts of electricity generation? Consider the amount of land that you would need with solar, what it's going to take to get all of that installed, and so on.
And by the way, there's a lot of stuff in China—if you look at all the big solar buildouts they did, they're ripping a lot of it up now. This is a longer conversation, but I really do question whether we can pin everything on solar. It's going to be a mix of stuff.
So your point may be right that, in the near term, to meet the current demand curve, nuclear is going to be economically challenged. But at some point here, there's an inflection that we have to meet.
So what's your most anticipated trend of 2026, Friedberg?
Iran becoming an independent democratic state. I think I'm just speaking generally about what's anticipated. A lot of people are anticipating that that's going to happen this year. There's an uprising in the streets, there's a weakening of the Ayatollahs, and there seems to be a moment underway.
Demographics are destiny. There's a lot of young people in Iran, and they do not want to live under the current rule. They want to be free.
But there is a major economic problem in Iran in terms of affordability. You think we have an affordability issue in the U.S.; in Iran, it's very hard for people to buy their basic necessities and meet their basic needs. That's why they're taking to the streets. There's a real economic crisis underway that's motivating this turnover.
Every year, everyone anticipates some big change in the Middle East, but this could be the biggest rewriting of the Middle East in a long time. Okay, Sacks, what do you have for the category, most anticipated trend of 2026?
I said auditing government spending at all levels. Decentralized DOGE.
That's a good one. That's a good one. Love it.
Yes. Let a thousand Nick Shirleys bloom.
Love it.
Let's do it. Audit everything. We need to normalize independent audits across the board. Whistleblowers, let's go.
It is not acceptable for Gavin Newsom, for example, to prohibit audits, as he did with homeless spending. All government spending needs to be opened up and audited by the public. We need people to see where it's going. That just has to happen.
The Pentagon—when are they going to pass an audit? That's one of our biggest line items. Let's get them audited. Hey, there's something President Trump could do.
They're actually at least trying.
They're at least failing the audit.
Gavin Newsom's prohibiting the audit.
Start where you're prohibiting them first. How about that? Yeah, I mean, just audit everything. I think that would be—I love it. What do you have, Chamath?
I have a corollary to Friedberg's, which is the expansion of this Trump doctrine. Independent of your politics, if you are an economic actor—you own a business, you invest in the stock market, whatever it is, you speculate in cryptocurrencies—you must understand the movements on the chessboard in 2026.
The best framework that I've used to organize myself is this idea of unilateralism and economic resilience. It's just a ginormous trend, and I think the output of it is going to be massive GDP prints on top of everything else.
Yeah. I think I'm going to stick with my prediction from last year going into 2026 again: the wrath of Lina Khan ending. We saw the M&A train start with xAI and X, obviously, Netflix and Warner Bros., and Google and Wiz. So many deals are ready to be done, and they're starting to pop off in M&A. We can debate what structure they are, regulators, and so on, but they're happening.
I'll go with IPOs coming back right now. You've got to anticipate that 2 of the following will file: SpaceX, Anduril, Stripe, Anthropic, and OpenAI. I think 2 of those file, and it's going to be gangbusters.
The public wants these shares. They're buying them in the secondary markets. We have half as many publicly traded companies, and the public would like to participate. This is something Trump can uniquely do that the Democrats were trying to stop and slow down, which was M&A and IPOs.
I think this will be the year of the mega-IPO, and it's going to be very exciting for Silicon Valley. It's going to be very exciting for the employees at these companies and for the pension funds and endowments that own shares in these companies. They're going to be able to take that money and put it to good use, hopefully. SpaceX, Anduril, Stripe, Anthropic, and OpenAI are on the short list of companies that could go public this year.
All right, a fun one we like to do is the most anticipated media. What are you looking forward to in 2026? Last year I said Superman and Andor season 2. Superman did great, and Andor did amazing. It's the best TV show of the 21st century.
I think Chamath said, “The enormity of the files that will be declassified.” The Epstein files—we're halfway there. The JFK files—we haven't seen those. Gavin said 1923 season 2. I don't even know what that is, but okay.
Taylor Sheridan. That's a Taylor Sheridan show.
Oh, that's the Taylor Sheridan one. Yeah, yeah. Are you watching Landman? Landman's pretty great.
Friedberg, you said AI video games. What do you have this year, Friedberg? You love the media. You're a cinephile. What are you looking forward to in media in 2026?
This isn't as much what I'm looking forward to, but I do think the big trend in media is going to be citizen journalism doing exposés. I think we're just at the beginning of the exposé and man-on-the-street trend.
Man-on-the-street reporting, pushing stuff, getting cameras in people's faces. The work of journalism has been decentralized, and I think there's going to be so much more that's shared and uncovered this year. Okay, do you have one?
And by the way, the difference between what Nick Shirley is doing and what we've seen in the past—and what I think is going to be the new trend—is that much of the citizen journalism in the past has been, to some degree, a little more passive. It's sort of, “Hey, I caught this thing and I observed it.” But now there are people who are going to actively take a camera and say, “I'm going to go discover this thing. I'm going to go deep on it.” That's what I think we're going to see happen in a big way this year.
Well, there's a monetization path. You have Substack, where people can give donations, GoFundMe, and, on top of that, YouTube and X allowing you to share revenue in this particular category, which they previously did. As Nick Shirley pointed out, that means there's a path to profitability: get more clicks, get more views, make more money, and then reinvest it. I think I like your choice a lot. Chamath, do you have something you're anticipating?
Exactly the same thing as Friedberg. I'll just double down.
Okay. And what about you, Sacks? Do you have something you want to double down on?
Investigative journalism.
No, I thought we were talking about entertainment here. Well, yeah, I thought—it's okay. These guys zigged where we zagged.
One of my weird things that I watch on TikTok, and TikTok just keeps showing them to me, is these auditing videos. Have you guys ever seen these First Amendment auditors? These are people who take a camera, stand on the street, point the camera into someone's store or a bank's window, and film the people in the bank.
Then they wind up spraying each other. [laughter] There are always people who come out and say, “You can't do that. You're not allowed to do that.” The auditor says, “Okay, I would like you to leave. I don't want to talk to you.” They just do this and instigate people to call the police.
What they're doing is auditing whether the police understand First Amendment rights. How the police react is basically the end of the video. Sometimes the police say, “You can't do that,” and the auditor says, “Yes, I can. Call a supervisor.” Then they teach the police officer that you're allowed to stand in public places and film. Other times, they go to the business owner and say, “This guy's allowed to do this. Leave him alone.” I don't know why, but these videos are so entertaining. I love watching them. [laughter]
It's so good. You never know what's going to happen. Each one is a whole new adventure. I don't know why it's such an interesting form of content. I love watching it.
It's a uniquely American phenomenon: establishing your freedom and your First Amendment rights. The stoner comes out sometimes and tries to physically confront the guy, but then you'll have a woman walking with her 2-year-old child and she'll get into it and say, “Hey, don't violate his constitutional rights. He's allowed to be.” That's the best. They even go into the really dicey ones, where they go to the parking lot of a prison or into the lobby of a police station and do it. It gets pretty spicy. What do you have, Sacks? Do you have any media you're looking forward to? These guys are going with First Amendment investigative journalism, yada yada. What do you have, Sacks?
Well, the new Christopher Nolan movie is coming out. The Odyssey looks interesting.
Great call. Great call.
Yeah, that's mine. Great call. I went with The Odyssey.
What is that about?
The Odyssey.
Oh, come on.
No. Homer's Odyssey.
Yes, but as interpreted by the great director of our time, Christopher Nolan.
Have any of you idiots actually read The Odyssey? This is not great. I'm sorry, but The Odyssey is a terrible book.
All right. Okay, so we're going to get some good comments there.
You guys are all such wannabe poser intellectuals.
Christopher Nolan is just great. It's IMAX. It's going to be epic.
It's trash. That book is blah.
Okay. You don't like iambic pentameter.
So the idea of some confused person making a movie about it is also—how do I short that? Okay, that is actually—can I short that?
Can I short that movie on Polymarket? I'm sure there will be. You could probably bet on the box office. Pull it up, Nick.
I bet zero.
Okay, it's going to be a lot more than zero, I'm sure. I'm also a big fan of Timothée Chalamet, and Dune: Part Three is coming out. I also like Avengers: Doomsday.
Dune 3.
Oh, I love Dune.
Doom Part Three.
I love Doom 1 and 2 so much.
2 was great. I think 1 was a little bit of—
Doomsday will be good. Doomsday, actually, I go with that. Avengers: Doomsday is going to be, I think, fantastic. Robert Downey Jr. as Doctor Doom, setting up Secret Wars, and then you're going to tie up all the previous Marvel strings.
Look at The Odyssey: zero.
They're going to use Doomsday to bring back all the characters they killed.
Hey, besties. Here's a little something: I think we're about to hit 1 million YouTube subscribers before we do our Netflix deal and take the show off of YouTube. We hit a million before we—
Yeah. Wasn't it supposed to happen like a year and a half ago? We're going to have like a million-subscriber party a year and a half ago. You know why?
Ask me. Ask me what I do.
Tell me why. Tell me why. Sacks.
Because we never tell our subscribers to hit the like button or smash the subscribe button, whatever.
Yeah. Every other podcast I watch on YouTube, and I watch them all, they're always like, “Hit the like button. Hit the like button.”
Oh, God. Well, you know what too is the big trend now is to just do 10-minute quick hits as news breaks. So this is people flooding the channel.
Ask me what I think about getting to a million subscribers.
Any thoughts on this seminal moment for the All-In Pod hitting 1 million subscribers on YouTube? Go ahead. I'm sorry. 1 million.
You're welcome. [laughter] You're welcome. You're welcome.
What's this week at?
Oh, This Week in Startups is a quarter million. Don't worry about it. It's a niche show. Don't worry about it. Not bad. Just a frag. Not bad. Not bad.
Hey, listen. I love doing it. The reason there's four of us—
There's four of us, one of you. So we have four times the subscribers.
Yeah. Something like that. Yeah. They listen. One of the great things you learn in media is when you build a super team, a super band, it can actually do better than everybody individually. So collectively, we can do better. And listen, it's not for me to say.
Listen, I think you've adjusted to being Ringo Starr. [laughter] Everybody says the same thing to me. Without the show, this would be propaganda and it would be Trump's personal—
Who says that to you?
Everybody wants to walk Moose. Everybody says the drink. I'm the stir in the drink. I'm the one who stirs the drink.
You're keeping the private-equity wives watching. Great job.
Absolutely. I'm keeping them in. All the private equities love it. For every left winger that you keep watching the pod, we probably lose five MAGA people.
The MAGA people love hating me. They love hating me and they come back every week to hate-watch me and my takes.
Hate-watch.
They do. They hate-watch you, Chamath, and they watch for a reason. They tune in to hate you and hate me.
Well, listen, it's been a great year. Great job, besties. We kept the band together for one more year. Let's do a Polymarket: chances the All-In Podcast makes it to 2027.
Honestly, it's not 100%.
It's never 100%. We'll do the best we can. You guys love the show, smash the like button, comment, link, subscribe, whatever you want to do. Write a review. Tell everybody how much you love Chamath's sweaters, and we'll see you next week on another amazing All-In.
Bye-bye. [music]
Let your winners ride. We open-sourced it to the fans and they've just gone crazy with it. [music]
Besties are gone. That is my dog taking on a shared driveway. [music]
Oh man, my haberdasher will meet up. [music]
We should all just get a room and just have one big huge orgy because they're all just useless. It's like this sexual tension that we just need to release somehow. [laughter]
We need to get merch. [music]
I'm going all in.