Overview of Big Tech lost big in the 2026 election
This Vergecast episode breaks down how a major Democratic election wave became a referendum on big tech, AI, data centers, and corruption. Nilay Patel and David Pierce argue that the results showed broad public backlash against the tech industry’s political entanglements, AI hype, and the real-world costs of data centers, while also touching on the state of social media, OpenAI vs. Google, and a lightning round of product and policy news.
Election Takeaways: Anti-Corruption, Anti–Big Tech, Anti–Data Center
The hosts frame the election as a decisive loss for Republicans and for the tech companies aligned with them.
Core themes from the results
- Voters appeared to reject overt corruption and cronyism tied to tech CEOs and the Trump orbit.
- Data centers became a major local political issue, with 41 ballot measures in 10 states and all but one going against data centers.
- The election suggested that public skepticism toward AI is now politically actionable, not just abstract.
Why the backlash landed
- The hosts argue that people are being asked to absorb real costs:
- higher energy use,
- local infrastructure strain,
- noise and environmental impact from data centers,
- and the political influence of wealthy tech leaders.
- In exchange, the public sees products that often feel like “slop” rather than meaningful progress.
Main takeaway
The election signaled that tech companies can no longer rely on political influence alone; they need products that clearly improve people’s lives.
AI, Regulation, and the End of “Just Trust Us”
A major thread in the episode is that AI companies are running into a credibility wall.
What the hosts think changed
- AI is now widely associated with:
- corporate overreach,
- bad products,
- and huge infrastructure costs.
- The industry’s political posture is hurting it:
- CEOs are seen as close to Trump,
- tied to corruption narratives,
- and overly eager to push AI into everything.
Why the products are a problem
- The hosts repeatedly argue that AI products still are not good enough to justify their costs.
- They point to:
- weak consumer utility,
- unstable agent behavior,
- and unclear business models.
- Their bottom line: AI companies will not escape regulatory pressure until the products themselves become obviously better.
Political outlook
- They expect:
- more state-level privacy and surveillance laws,
- more lobbying battles over data centers,
- and a wave of congressional investigations.
- Democratic lawmakers are already signaling aggressive oversight.
Social Media, X, and Political Communication
The episode also examines how political discourse is being shaped by social platforms after the election.
X is no longer the center of political reality
- The hosts argue that X/Twitter has become a smaller, less influential bubble.
- Elon Musk’s ownership is framed as a failure:
- it didn’t make X bigger,
- didn’t make it more relevant,
- and didn’t produce the political influence he seemingly wanted.
Where political energy is now
- The more meaningful conversations seem to be happening on:
- Instagram,
- TikTok,
- Reddit,
- Threads,
- and in comment sections.
- Politicians are increasingly using short video hype-reel content instead of text posts.
The new style of politics
- Younger Democrats are speaking in internet-native formats.
- The hosts note that some established politicians are bad at this style and shouldn’t force it.
OpenAI, Google, and the Economics of AI Search
A large part of the show is spent on the fight between OpenAI and Google.
What OpenAI claims
- OpenAI says ChatGPT is increasingly functioning like a search engine and growing quickly.
- The hosts are skeptical of the way these numbers are presented, but agree that OpenAI is trying to position itself as a real competitor to Google.
Why Google is vulnerable
- Google is trying to replace its own search model with AI overviews and AI mode.
- The hosts argue this is dangerous because:
- AI search doesn’t monetize as well as traditional search,
- and Google may be cannibalizing its strongest business.
Big picture
- Google’s edge was always:
- user click data,
- strong ranking,
- and web traffic distribution.
- AI threatens that model by:
- reducing clicks,
- weakening the open web,
- and making the value chain less clear.
AI Agents, Spending, and Reality Checks
The episode pushes back hard on the idea that AI agents are already transformative.
The numbers don’t look impressive
- Uber reportedly fulfilled only four rides from AI agents in a quarter.
- DoorDash and Uber Eats numbers were similarly tiny.
- The hosts view these as proof that agents are not yet a major consumer behavior.
Enterprises are also slowing down
- Companies are starting to cap or reduce AI usage because the cost isn’t justified.
- Example cited:
- Apple reportedly lowered daily internal token limits.
- The hosts suggest businesses are realizing that AI productivity gains are not yet worth the expense.
Their broader point
- The real money in AI is still mostly speculative or circular.
- Revenue is growing, but the hosts think much of it is coming from business spending that may not last.
Sam Altman, RAM, and the Cloud vs. Desktop Debate
One of the funniest side discussions centers on Sam Altman’s comments about RAM.
What Altman said
- Altman argued that it’s wasteful for people to keep a lot of RAM in their own computers if it sits idle most of the time.
- The hosts interpret this as part of OpenAI’s broader belief that compute should live in the cloud.
Why it landed badly
- The comment was read as:
- “I should have your RAM, not you.”
- The hosts see it as emblematic of OpenAI’s worldview:
- centralize compute,
- maximize utilization in data centers,
- and diminish the user’s control.
Their reaction
- They think the framing was tactically awful, even if the underlying point was about infrastructure efficiency.
- It became an easy meme because it sounded anti-consumer.
Brendan Carr Watch
The recurring “Brendan Carr is a Dummy” segment gets a brief update.
What happened
- Carr has been threatening broadcasters and talking tough about regulation.
- After the election, Democrats started openly discussing more aggressive oversight of the FCC and Carr himself.
The mood
- The hosts treat Carr as a politically vulnerable and relatively unserious figure.
- They joke that he may eventually be forced out or pushed toward a job at Starlink.
Lightning Round Highlights
The episode closes with several quick hits.
New MacBook Pro with touchscreen
- Nilay reviews Apple’s new touchscreen MacBook Pro and says:
- the touchscreen works fine,
- but is not a major reason to buy the laptop,
- and is easy to forget about.
- He does note:
- the machine is lighter and thinner,
- performance is excellent,
- but the speakers are somewhat worse than before.
SpaceX and Boost Mobile
- SpaceX is reportedly exploring a deal involving Boost Mobile.
- The hosts see this as another example of Elon Musk trying to turn ambition into telecom power.
- They doubt SpaceX is properly positioned to become a true fourth wireless carrier.
The “what is a photo” AI controversy
- A reporter used ChatGPT to brighten a photo and got accused of altering it in a misleading way.
- The hosts explain the key difference:
- traditional photo editing changes brightness/contrast deterministically,
- while ChatGPT-generated edits can create a different image entirely.
- Their term for the confusion: computational relativism.
Final Takeaway
The episode’s central argument is that the tech industry is entering a backlash phase:
- voters are rejecting corruption and influence-peddling,
- AI is losing the benefit of the doubt,
- data centers are becoming politically toxic,
- and tech companies can’t rely on hype alone.
The hosts’ bottom line is simple: if AI and big tech want to survive the next political era, they need products that are visibly useful, not just expensive, powerful, or influential.