Overview of Pivot from New York Magazine
This episode of Pivot centers on the growing tension between Big Tech’s immense power and the lack of meaningful regulation around AI, elections, and political influence. Kara Swisher and Scott Galloway break down NVIDIA’s massive stock buyback and new AI “guardrails” tool, the White House’s cozy relationship with tech leaders during President Trump’s China visit, and Elon Musk’s voter-data strategy disguised as election information. They also touch on ad platform hypocrisy around Musk’s documentary, and close with their weekly wins and fails, ranging from steel tariffs to historians, the Pope, and the Cornell fraternity case.
NVIDIA, AI Guardrails, and the “Arms Dealer” Problem
Massive stock buyback
- NVIDIA announced the largest stock buyback in U.S. history, expanding its repurchase program by $150 billion for a total authorization of $235 billion.
- Scott argues buybacks often signal that a company has run out of productive ideas for capital deployment.
- He frames the move as both a corporate signal and a macroeconomic symptom: capital increasingly flowing to shareholders rather than workers or consumers.
AI safety tools and self-regulation
- NVIDIA also announced a tool to monitor and prevent AI agents from “going rogue.”
- Kara and Scott see the timing as revealing: AI companies are trying to market safety products while resisting external regulation.
- Their core critique: the companies creating the risk should not be trusted to police it alone.
- Scott’s metaphor: “NVIDIA selling guardrails is like an arms dealer selling bulletproof vests.”
Key takeaway
- They argue the industry is moving toward a world where AI systems are powerful enough to cause harm, while governance remains weak, inconsistent, and largely self-imposed.
Big Tech at the White House and the China AI Stalemate
Tech leaders at the table
- The hosts discuss the White House dinner with tech leaders including Jeff Bezos, Sam Altman, Mark Zuckerberg, Elon Musk, Jensen Huang, and Tim Cook.
- Scott sees this as evidence that the same small group of tech executives is repeatedly given outsized access to political power.
Trump’s incentive to protect AI growth
- Scott argues Trump is reluctant to regulate AI because:
- U.S. GDP growth is heavily driven by tech capex.
- Hyperscalers and AI-related investment are among the only bright spots in the economy.
- In his view, Trump sees AI as too important to disrupt, especially heading into an election.
China visit: all optics, no substance
- The Trump-Xi meeting produced little beyond symbolism — notably, the “two pandas.”
- Kara says the summit was a missed opportunity to establish AI cooperation or shared guardrails with China.
- She argues a serious AI framework could have served both diplomacy and domestic politics.
Core criticism
- The episode portrays the situation as courtship plus coercion:
- tech leaders are invited in, monitored, and leveraged,
- but no durable policy emerges.
Elon Musk, Voter Data, and Election Manipulation
VoteSafe.org concerns
- The hosts warn about votesafe.org, a site tied to Elon Musk’s America PAC.
- The site can collect:
- name
- address
- phone number
- political information
- browsing activity
- Its privacy policy says the data may be shared or sold to business partners.
Their view
- Scott calls it “data harvesting posing as civic service.”
- Kara says it is misleading and should be illegal if it isn’t already.
- Both emphasize that Musk is not just “influencing” voters — he is building a political list.
Broader concern
- They frame this as part of a larger erosion of democratic norms:
- voter suppression
- misinformation
- gerrymandering
- concentration of political power in a few wealthy hands
Ad Platforms and the Elon Musk Documentary
The controversy
- Meta and YouTube initially refused to run ads for Alex Gibney’s upcoming Elon Musk documentary.
- Both later reversed course, calling the rejection an error.
- TikTok still reportedly would not run the ads; X obviously did not.
Scott’s marketing take
- He recommends leaning into the backlash:
- “Banned by Big Tech” can be a strong marketing angle.
- Use billboards near Tesla dealerships.
- Promote on skeptical platforms like Bluesky, Threads, Reddit, podcasts, and theater pre-roll.
Bigger point
- The reversal reinforces the idea that major platforms apply different standards when criticism targets billionaires.
SNL, AI Fear, and the “Doomer” Narrative
- The episode references an SNL sketch parodying Anthropic CEO Dario Amodei and the AI industry’s doom warnings.
- Scott and Kara agree the satire lands because the CEOs sound like they are both warning about and selling the danger.
- They reject the idea that AI agents are simply “going rogue” on their own; in their view, the humans who designed them bear responsibility.
Wins and Fails
Scott’s fail: steel mill “performative masculinity”
- He criticizes a heavily subsidized steel plant in Iowa.
- He argues it cost taxpayers about $9 million per job.
- He calls it performative testosterone: expensive, symbolic industrial politics that may destroy more jobs than it creates through downstream costs and tariffs.
Scott’s win: historians as public intellectuals
- He praises historians like Timothy Snyder, Anne Applebaum, and Heather Cox Richardson.
- His takeaway:
- they are courageous,
- they communicate clearly,
- and they are using their platforms to shape the political moment.
- He highlights their warnings about:
- misplaced war spending,
- support for Ukraine,
- and missed opportunities to invest in renewables.
Kara’s win: the Pope
- She praises the Pope for:
- kindness,
- moral clarity,
- drawing huge crowds,
- and speaking meaningfully on war and AI.
- She sees him as a positive, grounding public voice.
Kara’s fail: the Cornell fraternity case
- She discusses a disturbing Cornell sexual assault case and argues the school and local authorities may have failed to conduct a proper investigation.
- She says the situation reflects a broader anger among women over how these cases are handled.
- Scott agrees campuses often minimize incidents to protect brand reputation, which is exactly why transparent investigations matter.
Closing Thought
The episode’s throughline is accountability: in AI, in politics, in media, and in universities. Kara and Scott argue that the people creating the biggest risks — from AI labs to billionaires to institutions protecting their own brands — are often the least trustworthy to regulate themselves. Their recurring message: we need real guardrails, not PR guardrails.