FH For Humanity: An AI Risk Podcast

Podcast

For Humanity: An AI Risk Podcast

For Humanity, An AI Risk Podcast is the the AI Risk Podcast for regular people. Peabody, duPont-Columbia and multi-Emmy Award-winning former journalist John Sherman explores the shocking worst-case scenario of artificial intelligence: human extinction. The makers of AI openly admit it their work could kill all humans, in as soon as 2-10 years. This podcast is solely about the threat of human extinction from AGI. We’ll name and meet the heroes and villains, explore the issues and ideas, and what you can do to help save humanity. theairisknetwork.substack.com

Website ↗ Last published 29 October 2026 148 posts collected by Sputnik

Said on For Humanity: An AI Risk Podcast

“Those are the first jobs. Like the coders wrote themselves out of their own jobs knowingly. It's like going to work every day and cutting off a little piece of your arm and eating it until you have no arm left. And it's like, you were eating your arm. How did you think you would have an arm when you kept eating your arm?” — John Sherman, For Humanity: An AI Risk Podcast
John Sherman · 29 Oct 2026 Listen →
“Our history is littered with, I mean, it's just full of examples, after example, where companies were left alone to police themselves or regulate themselves. And they will basically do anything and everything, it seems like almost all the time, until they were physically stopped from doing it. And you'd like to give them the benefit of the doubt. But I mean, history shows that normally that doesn't go well.” — Dewayne Moore, For Humanity: An AI Risk Podcast
Dewayne Moore · 29 Oct 2026 Listen →
“If you say you're worried what you're making is going to kill everybody, like everybody, and you're serious, there's no excuse to keep doing it in my mind. I don't care if, well, if we don't, then they'll do it. Okay. So just stop. Stop and go fly to the other companies doing it and spend all your time convincing them to stop, you know, if you're serious. I think that's a cop out.” — Dewayne Moore, For Humanity: An AI Risk Podcast
Dewayne Moore · 29 Oct 2026 Listen →
“I suspect he might have a slight economic dependence on this not slowing down, or arguing against data centers. Slight bias.” — Roman Yampolskiy, For Humanity: An AI Risk Podcast
Roman Yampolskiy · 25 Sep 2026 Listen →
“…Sometimes the safest thing to do is let it burn.” — Gary Oldham, For Humanity: An AI Risk Podcast
Gary Oldham · 29 Aug 2026 Listen →

Hosts

For Humanity: An AI Risk Podcast is hosted by John Sherman and Michael Zafiris .

Recent Episodes

  1. 29 Oct 2026

    The People Building AI Say They're Worried. One New Listener Noticed.

    With Dewayne Moore

  2. 25 Sep 2026

    Tools, Not Replacements: Roman Yampolskiy's Line in the Sand

    With Roman Yampolskiy

  3. 15 Sep 2026

    The AI Kill Switch That Still Doesn't Exist

    With Jim VandeHei

  4. 3 Sep 2026

    The Volunteer Fire Department's Plan Is To Watch It Burn

    With David Senter

  5. 29 Aug 2026

    The Data Center Fire Risk Nobody's Talking About | For Humanity #91

    With Gary Oldham

  6. 13 Jun 2026

    The fear inside the AI labs is real, and it keeps getting harder to ignore

    With Jeffrey Ladish

  7. 30 May 2026

    The AI Buildout Has a Physical Speed Limit

    With John Billow

  8. 13 May 2026

    The Week the Spreadsheet Won

  9. 4 May 2026

    What this network is, and what it isn't

  10. 2 May 2026

    She Spent 12 Years Fighting Amazon. Now She Wants to Cut the Power to AI.

    With Connor Axiotis and Elena Schlossberg

  11. 26 Apr 2026

    What the Data Actually Says About AI Safety Right Now

  12. 22 Apr 2026

    We’re Growing - And Here’s Why That Matters Right Now

Guests

Roman Yampolskiy (7 episodes), Liron Shapira (6 episodes), Eduard Harris (4 episodes), Holly Elmore (4 episodes), Mike Brooks (2 episodes), Chris Gerrby (2 episodes), Dustin Burham (2 episodes), Jon Dodd (2 episodes), Kat Woods (2 episodes), Louis Berman (2 episodes), Max Winga (2 episodes), Timothy Lee (2 episodes), and 55 more.