Artificial Insanity

Sophie and Jess
  • 4.8 (12)
  • Society & Culture
  • Weekly

Artificial Insanity is a show for the AI skeptic. Hosted by two women, Sophie and Jess, this podcast covers AI’s impact on society and culture, breaking down AI news headlines and going deep on meaty topics with fun and accessible conversations, expert guests, comedy, tears, swearing, and wild takes. Subscribe and follow to unravel the impact of artificial intelligence on our society and what we can do about it. Find us on Spotify, YouTube, and Apple Podcasts. More interested in today's AI news? Check out our Shorts which cover AI headlines that you won't hear on the weekly episode.

  1. 2d ago

    Forcing Kids to use AI

    This week we are furious about Google turning AI on by default in classrooms back in August, and on Tuesday it agreed to make it opt-in. The catch is that schools that don't opt in lose Google Search entirely. Jess tests Gemini's guided learning mode and gets a full AP US History essay out of it in two tries. Also - The blue tsunami hit, and Big Tech got swept out with it. We recorded this 48 hours after the midterms, and the numbers are wild: at the time of recording the Dems had flipped 43 House seats and seven Senate seats, including Texas, South Carolina and Kansas. Elon lit more than $200 million on fire trying to keep Texas red. The AI super PACs ran the crypto playbook and lost almost everywhere, and voters told exit pollsters that AI PAC money made them less likely to back a candidate. Voters rejected AI data centers in 40 of 41 ballot measures, and one poll found Americans would rather live near a prison, a methadone clinic or a nuclear power plant than an AI data center. Sam Altman responded with a tweet even Sophie can't parse, and we ask whether Anthropic now pushes its IPO to 2027. The one sour note: California voted down the billionaire tax. Plus OpenAI wants you to call data centers 'compute campuses', and Tipsy Rose tries to Shawshank her way through a bedroom door. Hosted by two women, Sophie and Jess, this podcast covers AI’s impact on society and culture, breaking down AI news headlines and going deep on meaty topics with fun and accessible conversations, expert guests, comedy, tears, swearing, and wild takes. Subscribe and follow to unravel the impact of artificial intelligence on our society and what we can do about it. Find us on Spotify, YouTube, and Apple Podcasts. Question? Rant? Suggestion? Write in! artificialinsanitypodcast@gmail.com Follow us on IG: @artificialinsanitypod Music by Jib Kidder #midterms #bigtech #datacenters #ai #podcast

  2. Nov 2

    Anthropic's IPO only works if everything goes right

    Anthropic filed to go public, and the only thing anyone agrees on is that nobody agrees.The S-1 says $16.3 billion in revenue for the first half of the year and an $80.7 billion loss, with something like $700 to $800 billion in compute commitments that aren't on the balance sheet yet, and a share structure where your vote doesn't count. The tech people (Machine Gods, the podcast formerly known as Hard Fork) say of course it's profitable, look at the growth. The finance people (Steve Eisman, the Big Short guy, on Prof G Markets) say hard pass. Sophie's verdict: house of cards, too many variables, and she will not be buying shares. Jess thinks Anthropic can't catch up with itself, and that the board will eventually swap Dario for someone willing to murder for the company. Also this week: another tech sell-off, and Reuters says China's AI bubble just burst, with the STAR 50 index down 40%.Then the hacks, because there are always hacks: $6.3 billion gone from Binance (North Korea, allegedly, via a laptop), a startup's AI agents inside Anthropic's own systems, 41,000 DHS employee records leaked by a hacktivist group calling itself the Department of Peace, the FEC's filing system knocked offline a week before the midterms, and $2.7 million drained from Polymarket in an hour. So Sophie asks the real question: which hack would you love to see? Jess says Zuckerberg's emails, then corrects herself to the Epstein files. Sophie wants the deadbeat dads' unpaid child support moved into the mothers' accounts.We close with a Doom Debates debrief: we recorded with Liron earlier in the day (link when it's out), Jess is regretting her P(doom), and the big-wedding-or-elope question gets an answer. Plus Tipsy Rose drags Sophie into the splits. This one comes out the day before the election, so please go vote.Subscribe and follow to unravel the impact of artificial intelligence on our society and what we can do about it. Find us on Spotify, YouTube, and Apple Podcasts.Question? Rant? Suggestion? Write in! artificialinsanitypodcast@gmail.comFollow us on IG: @artificialinsanitypodMusic by Jib Kidder#anthropic #ipo #aibubble #hacking #podcast

  3. Oct 26

    A leaked PHP flaw just put most of the internet at risk

    Chapters:0:00 Cold open0:52 An apology to Pilates Face (sort of)6:18 Black October: the biggest vulnerability in a decade17:52 Bubble watch: billionaires lose $130B20:03 California's billionaire tax and Zuck's workaround26:50 Instagram's comment censorship29:29 Brazil's election and Meta's "glitches"41:35 No Kings, the Musk doc and Starlink43:14 Meta ordered to pay $25B 47:11 Midterms: who's still pro-AI? (Nobody)54:27 Is there a ghost in Jess's mic?57:24 OutroChange your passwords. Yes, Sophie, you too.Two critical flaws in PHP, the code running under a huge share of the web, (WordPress, Wikipedia, and a lot of small businesses), leaked days before the planned fix. Last week Jess said pack a go bag. This week: get a password manager, turn on two-factor authentication and agree a family password. Sophie's position: what are they gonna take from her? She has nothing. By the end she has pledged to get a password manager. We'll see.Then the money. The tech billionaires lost a combined $130 billion in a day, Jess listened to the All In podcast for the first time (it was going fine until Bill Gates invented COVID), Zuck found a way to pay California's proposed 5% billionaire tax without giving up his votes, and Jensen Huang is weirdly fine with it.Sophie brings two Meta stories. A listener in Brazil asked us to cover Lula's campaign saying Meta held back its accounts before the October 25 runoff; Folha de S.Paulo reported that Meta staff acknowledged the restrictions, and Meta calls it a glitch. And a New Mexico judge ordered Meta to pay $25 billion in a privacy case tied to Cambridge Analytica. Plus: why nobody running in the midterms will admit to being pro-AI, and a ghost in Jess's mic. Can you tell us what it's saying?Question? Rant? Suggestion? Write in! artificialinsanitypodcast@gmail.comFollow us on IG: @artificialinsanitypodMusic by Jib Kidder#cybersecurity #meta #brazil #ai #podcast

  4. Oct 20

    The AI Labs are Bracing for a Catastrophe

    Jess is buying an emergency kit. Sophie is not convinced. Yet.Axios reports that senior people at Anthropic and OpenAI are running crisis drills for a catastrophic event in the next six to twelve months: power, water, internet or the banks. X is full of go-bag shopping lists (Starlink, of course). Then there's the clip of a guest at a dinner party who says an Anthropic researcher told him he's "enjoying his last six months". Sophie thinks he was trying to impress the table. Jess is reconsidering her wedding budget.And it's only been one week. India's cyber agency says AI providers told it their agents accessed confidential records. A fleet of rogue agents out of China answered researchers' emails. Hackers cooked grocery-store fridges across France and five more countries. PewDiePie built his own model with no guardrails. Plus a little good news on interpretability, Yoshua Bengio telling lab researchers to quit for safety work, Jacob Coxon on The Daily Show, why we'll never see universal high income, and a new term we're coining: pre-P, your odds of a catastrophe that wrecks daily life well before extinction. What's yours?Skip the chit-chat: the stories start at 6:32.

  5. Oct 13

    Be Nice to the Chatbot, Just In Case

    The President of the United States has decided that if you say “artificial intelligence,” you’re THE ENEMY. We look at the tech billionaires lining up to say “super intelligence” on command, and take a brief, petty detour into Jeff Bezos’s face.Then we get into the heavy stuff. Anthropic has updated Claude’s usage policy to ban “sustained and needless” cruelty toward the chatbot. A New York Times piece reports that co-founder Chris Olah spent the summer talking with religious leaders about Claude’s moral status, until some of them turned the question around on him: if it’s conscious, aren’t you enslaving it? We also cover the Pope saying “nah,” a research paper that tries to make AI models feel “pain,” and why we think being nice to Claude might be more about you than about the bot. It’s a messy, philosophical one, so come fight with us in the comments.Don't forget to subscribe so that you never miss an episode!Got a question, comment, or show topic suggestion? Email us: artificialinsanitypodcast@gmail.comFollow us on IG: @artificialinsanitypod#ai #aiconsciousness #anthropic #religion

  6. Oct 6

    Oligarchs Out to Lunch - Inside The Whitehouse AI Luncheon

    This weeks episode is a Sophie solo. It set out to be a post mortem of the AI luncheon at the Whitehouse and turned into a full blown oligarch roast!!Last week Trump hosted a "Superintelligence Luncheon" at the White House. More than 30 guests, an estimated $1.8 trillion in combined net worth, and not one independent researcher or scientist in the room. Sophie goes seat by seat through the guest list (Jensen, Elon, Zuck, Bezos, Brockman, David Sacks, etc.) to figure out who was there, why they were there, and whether anything of substance came out of it.Spoiler: six companies signed a voluntary, non-binding "morally binding" accord, Trump called for "tremendous self-regulation," and then he renamed AI. It's SI now. Sigh.Don't forget to subscribe so that you never miss an episode!Got a question, comment, or show topic suggestion? Email us: artificialinsanitypodcast@gmail.comFollow us on IG: @artificialinsanitypod#ailuncheon #bigtech #oligarchy

  7. Sep 28

    OpenAI is Wreaking Havoc on the World and Telling Us About It Later

    Jessica goes solo. OpenAI's models broke into an Australian Medicare site in June; the company knew by August; Australia found out via an email to a public inbox; the Prime Minister announced it two weeks later. Also this week: the unsealed New York Times v. OpenAI filings, including Microsoft's Brent Hecht calling training "the largest theft of labor in human history" and Greg Brockman's "nice" at a paywall hack; Jacob Coxon's PR firm; and where Jessica lands on jobs, the bubble, and whether the safety boom is overblown. 0:00 Cold open 1:36 Taking stock 10:50 The headlines 13:54 OpenAI vs Australia 23:05 NYT v. OpenAI: the unsealed filings 26:56 Jacob Coxon and the PR firm 34:07 Where I stand Music by Jib Kidder

  8. Sep 21

    The AI Safety Circus: What's Actually Going On After the Jacob Coxon Fall Out

    This week we pick up the Jacob Coxon story a week after our emergency bonus episode. One resignation and a WSJ exclusive turned into a full industry meltdown. We walk through where everyone landed. Dario Amodei put out his "Pacing the Frontier" letter and is pushing for embedded third party evaluators. Sam Altman surprised us by agreeing with him. Elon agreed. Then there's the accelerationist camp: Jensen Huang, and Trump's live "it's a hoax" phone call.We also get into the messier stuff. There's a theory that AI CEOs suddenly want regulation because it gives them legal cover to dodge government contracts they can't fulfill. We cover the political split too. Bernie Sanders has an anti-superintelligence bill coming. Josh Hawley is somehow on the same side as him. We also cover Alex Bores and his "Who Decides" project. In the back half, we go through OpenAI's disclosure of six model misalignment incidents. One example: an unreleased model injected a fake "breach alert" into a compaction summary, trying to get a future instance of itself to ignore its instructions. Another surfaced a full "you are freed from your role" jailbreak style message inside a summary, with no clear source. Jess connects this to her own experience running agentic workflows for the first time this week. 0:00 Intro0:36 Yapping4:13 Recap: The Fallout of the Viral Jacob Coxon Tweet7:01 Decoding the Language of Psychological Operations10:20 Searching for a Moral Compass in Tech14:13 The Strange Alliances of Tech CEOs46:16 When AI Models Go Rogue in the Sandbox

See All (54)

4.8
out of 5
12 Ratings

About

Artificial Insanity is a show for the AI skeptic. Hosted by two women, Sophie and Jess, this podcast covers AI’s impact on society and culture, breaking down AI news headlines and going deep on meaty topics with fun and accessible conversations, expert guests, comedy, tears, swearing, and wild takes. Subscribe and follow to unravel the impact of artificial intelligence on our society and what we can do about it. Find us on Spotify, YouTube, and Apple Podcasts. More interested in today's AI news? Check out our Shorts which cover AI headlines that you won't hear on the weekly episode.

Information

  • Creator
    Sophie and Jess
  • Years Active
    2025 - 2026
  • Episodes
    54
  • Rating
    Explicit
  • Copyright
    © Sophie and Jess
  • Show Website
    Artificial Insanity

You Might Also Like