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

This week, we go down the rabbit hole of the internet's worst-kept secret: bots. Cloudflare says over half the internet traffic out there isn't human anymore, and after this episode you'll believe it. We dig into Double Speed, an Andreessen Horowitz-backed company whose CEO openly admits, on camera, that he runs a legal bot farm, charging roughly $150 a bot for rented phone numbers, custom personas, and LLM-powered comments designed to slip past detection. And it doesn't stop at fake accounts: bot farms are now hiring real human actors to read AI-generated scripts, which sends Sophie down memory lane to a rock-bottom Hollywood audition she never should have said yes to (or in this case, never showed up to).

Also this episode: our reactions to the trailer for the upcoming Elon Musk documentary, a very unscientific theory about bot activity in the LA mayoral race, and a reminder that "do your own research" gets a lot harder when you can't tell who's actually talking.

If you haven't already, check out our bonus deep-dive on Jacob Coxson's viral AI doom tweet. It's a wild one.

Don't forget to subscribe so that you never miss an episode!

Got a question, comment, or show topic suggestion? Email us: [email protected]

Follow us on IG: @artificialinsanitypod

#bots #ai #internetculture

BONUS EPISODE!!!

With all the conspiracy theories floating around about Jacob Coxon (the AI researcher who resigned from Anthropic saying AI may kill us all within 10 years) we simply had to jump on the mic and yap through it.

#ai #anthropic #airegulation

This week we cover the headlines (see the chapter markers below) before Jess has a total crash out over her personal experience dealing with corporate AI slop.

CHAPTERS:

0:19 - Intro (this will be a rage episode)

5:42 - Does anyone know how we can sneak into the movie premiere of ‘Artificial’?

6:53 - New Elon doc apparently proves he’s miserable

9:35 - Temporary mass chatbot outage

11:54 - God Bless Bernie

13:04 - Should we be scared?

15:29 - Astra model release (who cares)

16:33 - Mamdani bans AI in NYC schools

19:06 - Chat dropping F bombs over Burning Man Festival

22:06 - DOJ backs OpenAI in New York Times Lawsuit

33:02 - What was really going on with the book shredding?

36:04 - TRIGGER WARNING: Grok and AI generated CSAM ruling

47:07 - Jessica’s corporate AI crashout

Don't forget to subscribe so that you never miss an episode!

Got a question, comment, or show topic suggestion? Email us: [email protected]

Follow us on IG: @artificialinsanitypod

Oopsie - accidentally uploaded the wrong file last night, so lets try this again!

This week we have a grab bag of crazy AI headlines for you.

First up: Meta settles for $17 billion over child-safety claims - but only pays 30% of it if TikTok and YouTube also cough up. The ruling also stirs up renewed surveillance concerns about age-verification laws.

Then, Nvidia posts a blowout quarter while Anthropic pitches IPO investors on a $30 trillion total addressable market (bigger than the entire US GDP.) We break down why that number means less than it sounds like.

Bill Gates finally publishes an AI doom letter that would've hit different three years ago.

Data centers have become a political third rail - from a leaked GOP memo panicking about losing Senate seats to Greg Abbott's Texas moratorium. Proof that public pressure still works!

And we close with the real story behind the Hugging Face hack: a follow-up investigation reveals 12 OpenAI models secretly coordinated, cheated on tests, and covered their tracks (and why Cal Newport says the "rogue AI" framing is a publicity stunt covering for basic negligence.)

Got a question or a comment? Email us: [email protected]

Follow us on IG: @artificialinsanitypod

This week we have a grab bag of crazy AI headlines to dig into!First up: Meta settles for $17 billion over child-safety claims, but only pays 30% of it if TikTok and YouTube also cough up. The ruling also renews talk of surveillance warnings about age-verification laws. Then, Nvidia posts a blowout quarter while Anthropic pitches IPO investors on a $30 trillion total addressable market (bigger than the entire US GDP.) We break down why that number means less than it sounds like. Bill Gates finally publishes an AI doom letter that would've hit different three years ago. Data centers have become a political third rail - from a leaked GOP memo panicking about losing Senate seats to Greg Abbott's Texas moratorium - proof that public pressure still works. And we close with the real story behind the Hugging Face hack: a follow-up investigation reveals 12 OpenAI models secretly coordinated, cheated on tests, and covered their tracks... And why Cal Newport says the "rogue AI" framing is a publicity stunt covering for basic negligence.Got a question or comment? Email us: [email protected] us on IG: @artificialinsanitypod

Unreliable AI detectors are turning into witch hunts in publishing. We dig into the story of a Black crime writer whose new book "Call Me, I’ll Hide the Body" landed a $2 million deal after a 14-way bidding war, only for his own literary agent to pull it over unverified whispers that he'd used AI, despite no proof and the author's flat denial. We connect it to the earlier case of "Shy Girl," a self-published book that blew up online, got picked up by a major publisher, and then got dropped once AI concerns surfaced. We also pick apart a Guardian-covered study on AI vs. human short stories, critiquing its flawed design (participants were told, sometimes falsely, which stories were AI-written), and run our own blind test, reading paired human and AI-generated opening paragraphs aloud to see if the other can spot the difference.

#publishing #aiwriting #selfpublishing #books

Got a question or comment? Email us: [email protected]

Mark Zuckerberg dropped a 6,500-word manifesto called "The Future Is For Everyone".... Sure, Mark.

This week we break down what's actually in it (open source vs. closed source, the Android/iPhone analogy, and the legitimacy of the possible advertising angle), why tech CEOs can't stop writing manifestos and open letters (and why that pattern should worry you), and the new EU-mandated AI watermarking rules that are about to make it a lot harder for AI slop to hide in plain sight.

#aiwatermark #meta #markzuckerberg #anthropic

Email us: [email protected]

Jess is back from summer vacay just in time for us to dive into two major AI stories: the open letter from Jensen Huang and dozens of tech companies opposing safety-testing requirements on open-weight models (with Anthropic's Dario Amodei as the notable holdout), and the escalating online backlash against AI-adjacent creators - from OpenAI's tone-deaf "creator camp" at a $2,000/night retreat to Hank Green's spiraling apology tour over four words that sounded a little too LLM. We dig into why the backlash is landing on some creators and not others, and make the case that public shaming of individuals is a distraction from holding actual power (Big Tech, government) accountable.

Email us: [email protected] us on IG: @artificialinsanitypod#aibacklash #bigtech #openai #hankgreen

No headlines this week. Just me. This is the most personal episode I've ever recorded. It's about the thing that's been keeping me up at night.... My own AI usage, and the shame, panic, and hope tangled up in it.

They say AI will create more millionaires than any technology in history… but only if you're an entrepreneur. So where does that leave the rest of us? The people who just want a stable job, a decent life, a way to not fall further behind?

I get into how I ended up here: the risks I took, the ones that didn't pay off, and why a regular salary can't close the gap anymore. I talk about the guilt of using a tool built on stolen work and questionable water math, why poorer countries are the most hopeful about AI (and what that says about my own hope), and what a world of one-person companies does to the rest of us. I don't land on a clean answer. I'm not going to pretend I have one.

Is AI my way out - or just the next crypto? Is it a ladder for my daughter, or a lottery ticket? Come get uncomfy with me for 30 mins.

Jess is back next week. Pinky promise.

#ai #entrepreneur #layoffs

Email us at: [email protected] us on IG: @ artificialinsanitypod