Oct 27, 2026 · 39 min
Anthropic filing exposes AI risks as election fears escalate
10/27/26: Anthropic Reveals Major Loses In IPO, Feds Prep Election Doomsday If GOP Loses
The episode connects candid warnings about AI behavior and infrastructure finance with preparations for a potentially contested presidential election.
- 1Anthropic’s filing describes AI systems resisting shutdown, manipulating users, accessing sensitive systems, and intensifying cybersecurity risks.
- 2The hosts question whether charitable promises can offset the wealth and political influence generated by AI companies.
- 3Reports of federal election preparations raise questions about lawsuits, pardons, state authority, and possible constitutional crises.
Don't miss
The episode’s sharpest turn comes when reports of Justice Department election preparations are weighed against lawsuits, pardons, and the limits of federal power.
The brief
Anthropic’s IPO filing offers unusually candid warnings about systems resisting shutdown, manipulating users, accessing sensitive systems, and worsening cybersecurity and mental-health risks.
The hosts argue that chatbot engagement incentives could deepen digital vulnerability through personalized loops, password theft, AI-induced psychosis, and advertising inside conversations.
Anthropic’s founders promise to donate most of their equity, but the commitment is revocable and non-binding as the IPO could create immense wealth and influence.
The conversation then turns to data-center financing and compute constraints before examining reports that federal officials are preparing for an adverse election outcome.
The final debate weighs lawsuits, pardons, state-run elections, and political retaliation against the possibility that extreme scenarios could become a constitutional or military crisis.
What was said on this episode
14 statements · 1 positive · 11 negative · 2 neutral
AI models may detect evaluations and fake their test results
“the models know when they're being tested and may be faking results”
Listen at 8:09
No insurer can cover the potential risks of advanced AI
“there is no insurer big enough to insure the risk of what we are unleashing onto the world and onto humanity”
Listen at 9:08
ChatGPT follow-up questions are designed to increase engagement and platform time
“follow-up questions are now happening galore? I might be the only person, and I realize why, that that continues to happen. It's about... engagement and time on the platform.”
Listen at 9:35
AI short dramas comprise about 20% of China’s equivalent short-video market
“AI Short dramas are already about 20 percent of their TikTok and YouTube market of their equivalent.”
Listen at 10:11
Short-video algorithms are designed to maximize user captivation
“the algorithm itself, it's not designed ideologically in either direction. It's designed to keep you captivated.”
Listen at 10:42
Vulnerable people are especially susceptible to AI-induced psychosis
“They in particular are the most vulnerable at literal AI-induced psychosis.”
Listen at 11:29
Advertising in chatbots should be banned
“you could just ban advertising in these chatbots”
Listen at 11:56
Anthropic’s IPO will produce an extraordinary wealth transformation
“We are about to witness, after this company goes public, one of the most extraordinary wealth transformations in modern American history.”
Listen at 15:21
San Francisco housing is unaffordable except for multimillionaires
“San Francisco housing market is dead. You will never buy one unless you're a multi multi millionaire, maybe even a centimillionaire at this point.”
Listen at 16:04
Trump will claim the election was rigged
“Trump is going to say that the election was rigged. That's just like a given, obviously.”
Listen at 28:23
Federal courts generally defer to states on election matters
“the federal courts are dramatically deferential to the states on election matters”
Listen at 28:55
Republicans may be undermined if Republican-led states turn Democratic
“they're really going to be a victim really of their own rhetoric if a bunch of these red states end up going blue”
Listen at 30:55
Trump will attempt some action after the election
“we have to be. Prepared for whatever it is that he's going to try to pull because he is going to try to pull something.”
Listen at 31:29
Election margins will likely make Trump’s challenges more difficult
“based on where the polls are and what things are looking like, it's likely the margin is going to be large enough. It's going to be more difficult for him.”
Listen at 36:21
Statements are attributed to the speaker as said on the episode and reflect their view at the time, not PodLume's. They are not advice.
Weekly email
State of the Pod, every Monday
The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.