Featuring
Tucker Carlson sits in for Friedberg as the besties work through the Warner Bros. bidding war, Nick Fuentes's online rise, and the fear that AI will hollow out work. The sharpest sparks came when Jason insisted old news brands still carry political power and when Sacks challenged his five-year job-loss forecast. Sacks had the strongest episode: he kept the AI argument tied to measurable labor data while everyone else reached for the crystal ball.
Spice rack
Are CBS and CNN too irrelevant for concentrated ownership to carry political risk?
Original point: The Ellisons' support for Trump and expanding control of TikTok, CBS, and potentially CNN create at least the appearance of political favoritism, so Trump should stay out of the transaction.
What everyone argued
Chamath Palihapitiya
Chamath calls Jason's ownership concern a red herring because CBS and CNN have tiny audiences relative to modern platforms. Product quality, not the owner's identity, determines whether billions of people choose to watch, so the old networks would have to be rebuilt from scratch before their ownership mattered.
Jason Calacanis
Jason argues that CBS and CNN still reach millions and that combining them with TikTok-related influence under a family politically close to Trump creates a concentration problem. He distinguishes an actual quid pro quo, which he does not prove, from the appearance of one and says the president should avoid personal involvement.
Tucker Carlson
Tucker says the real speech threat is government or corporate censorship of YouTube, X, and Instagram, not who owns decaying television brands. If old media cannot command attention, changing its owner will not rebuild cultural authority; scrutiny should follow the platforms that control modern distribution.
Winner circle
Jason wins the narrow ownership-risk debate. CBS and CNN are weaker than YouTube or TikTok, but millions of viewers and control of major newsrooms are not political rounding errors. Hindsight strengthened his concern without proving his accusation: Paramount won the agreement and federal clearance, while states and critics kept contesting both concentration and political influence.
Commentary
Chamath Palihapitiya
Assumptions and fact checks
Who owns a news outlet does not materially change what audiences consume or believe if the product itself is unchanged.
Why it mattersOwnership can change leadership, budgets, distribution, editorial priorities, and which stories receive promotion. Audience choice constrains that power, but it does not erase it.
CBS and CNN are effectively unwatched and therefore irrelevant.
CheckThe language was hyperbolic, but the underlying factual premise is still wrong. Nielsen data reported in December 2025 put CBS Evening News alone at about 4.04 million nightly viewers; that is far below the leading broadcast newscast, not zero or immaterial reach.
Jason Calacanis
Jason wins by keeping two claims separate: the networks still have measurable reach, and political proximity creates a governance risk even short of proven corruption. The argument would be stronger with a precise theory of influence instead of repeating that it is 'undeniable.'
Assumptions and fact checks
The Ellisons' political support and media acquisitions create a meaningful appearance of quid pro quo even without proof of an explicit exchange.
Why it mattersThe appearance concern is reasonable when a politically connected bidder seeks regulatory clearance for assets that can shape news and distribution. It remains an appearance claim: DOJ says career staff ran an eight-month review, and no evidence cited here proves a corrupt exchange.
Millions of people still watch CBS news programming.
CheckCBS Evening News averaged about 4.04 million viewers a night in 2025, before counting the network's other news programs and digital reach.
Paramount was offering more money for all of WBD while Netflix's agreed transaction covered the studio and streaming assets.
CheckThe competing structures were accurately described in substance. Hindsight went further: WBD ultimately signed Paramount's improved all-cash deal at $31 per share, and shareholders approved it in April 2026.
Tucker Carlson
Tucker supplies the strongest missing comparison and the weakest direct rebuttal. 'There is a bigger threat' does not settle whether the smaller one is real.
Assumptions and fact checks
Censorship or control at large technology platforms is materially more consequential than ownership concentration in legacy television news.
Why it mattersAt national scale, platform ranking, access, and moderation can affect far more content and users than one television newsroom. That relative judgment does not make the smaller concentration harmless.
Will AI cause mass U.S. job displacement within five years, or will adaptation remain gradual?
Original point: AI and autonomous systems are already replacing entry-level service and driving work, and millions of jobs will disappear over the next two to five years unless policy deliberately slows or offsets the transition.
What everyone argued
Jason Calacanis
Jason says startup customers are buying AI agents specifically to avoid adding sales and support staff, while Waymo, automated drive-throughs, and Amazon's robotics plan show physical work moving the same way. He predicts millions of jobs will disappear within two to five years and argues that retraining, housing, health care, and education policy must get ahead of street protests.
David Sacks
Sacks argues that current layoff data show AI as a small stated cause, while Yale finds no discernible broad labor-market disruption. He expects a productivity boom with gradual occupational change, compares the transition to the decades-long shift from brick-and-mortar retail, and says the two-to-five-year mass-displacement case has not met its burden.
Winner circle
Sacks wins the evidence round; the five-year forecast stays open. He correctly distinguishes measured present effects from feared future ones, and his Challenger and Yale citations survive scrutiny. Jason identifies real substitution incentives, but overstates Waymo's share, turns Amazon's avoided future hires into eliminated jobs, and assigns AI more blame for youth unemployment than current research can support.
Commentary
Jason Calacanis
Jason sees the microeconomic incentive clearly: many buyers want the same output with fewer people. He loses discipline by counting planned hires that never occur as layoffs and by treating every weak entry-level statistic as an AI fingerprint.
Assumptions and fact checks
Rising unemployment among young people is already primarily caused by AI.
Why it mattersYouth labor outcomes weakened, but the New York Fed found the decline in AI-exposed postings began before ChatGPT and did not diverge between junior and senior roles. BLS data also show a mixed age pattern rather than a clean AI signature.
Millions of job losses and visible social protest will arrive within two to five years.
Why it mattersThe mechanism is plausible and the deadline has not passed. Current data do not support that scale yet, while rapid model and robotics improvements make confident dismissal premature.
Amazon expected automation to eliminate 600,000 jobs.
CheckThe reported internal target was to avoid more than 600,000 future U.S. hires by 2033 while handling roughly twice as many products, not to lay off 600,000 existing workers. The distinction matters to the timing and social impact Jason was arguing.
Waymo had already moved roughly one-third of rides in both San Francisco and Los Angeles to driverless vehicles.
CheckWaymo disclosed 15 million rides across all of 2025 and more than one million monthly rides by spring, but did not substantiate a one-third share in both cities. Independent estimates put San Francisco share around the low twenties and do not establish the same share for Los Angeles.
David Sacks
Sacks wins by guarding the calendar: evidence about today cannot prove the 2030 labor market, but it can defeat claims that mass displacement is already obvious. His weak spot is using aggregate calm to sound more certain about the next five years than the studies themselves are.
Assumptions and fact checks
The AI transition will resemble the decades-long shift to e-commerce more than a two-to-five-year labor shock.
Why it mattersHistory rewards caution about compressed technology timelines, but software agents can diffuse faster than physical retail infrastructure. Current evidence supports gradualism only so far, not for the full forecast horizon.
Aggregate labor data are sufficient to reject claims of meaningful current AI harm.
Why it mattersThey reject an economy-wide claim, not every narrow one. Yale explicitly notes that broad surveys may miss recent graduates or specific occupations, and the New York Fed likewise leaves room for limited effects.
November 2025 announced job cuts fell 53% from October; AI was cited for about 6,000 November cuts and roughly 4.7% of year-to-date cuts.
CheckChallenger reported 71,321 November cuts, down 53% month over month, with AI cited for 6,280 cuts that month and 54,694 through November—about 4.7% of the 1.17 million total announced cuts.
Yale Budget Lab found no discernible broad labor-market disruption in the first 33 months after ChatGPT's release.
CheckThat accurately summarizes Yale's December 2025 analysis. Its May 2026 follow-up still found no statistically or economically significant effect in average AI-exposed occupations, while emphasizing that aggregate data can miss narrow harms.
AI-linked investment accounted for about half of U.S. GDP growth in 2025.
CheckThe best official-style accounting is lower and more qualified: St. Louis Fed researchers estimate broad AI-related investment categories supplied 39% of real GDP growth through the first three quarters of 2025, falling from 30% in Q2 to 11% in Q3. Those categories also include software and R&D that are not entirely AI, so 'half' is too strong as an annual claim.
Chamath has the best demand-side question—will anyone watch?—but treats audience size as the only channel of influence. Editorial control over a few million nightly viewers remains a concrete mechanism, even if TikTok and YouTube are much larger.