Into The Machine with Tobias Rose-Stockwell

It has been a huge week for AI safety.
For those of us deeply concerned about this historically niche cause, it has felt a bit like one’s cult emo band has suddenly rocketed to the top of the charts. People suddenly care, a lot. And for good reason.
The principal reason is this tweet — the most viral tweet in recent memory — suggesting that the frontier labs do not have it figured out, and that we are on the edge of a dangerous precipice. (FWIW, I tend to agree with the tweet’s concerns.)
Calibrating risk is a difficult thing. We look for experts who can tell us how bad things are. With little evidence of our own we’re forced to look for proxies in knowledgable folks. In this case, the most knowledgable folks I know are desperate for a functioning governmental body that can effectively slow down the pace of AI development. No one has it figured out.
Beneath this, I am not convinced artificial minds will kill us imminently. But I do believe we need to have a clear and sober discourse on what types of AI alignment efforts may or may not be working.
In line with this, today I’m publishing a conversation on one (arguably very edgy) side of the debate — with Nate Soares, president of the Machine Intelligence Research Institute. Nate wrote the book, If Anyone Builds It, Everyone Dies, along with his coauthor Eliezer Yudkowsky. I wanted to probe the dimensions of his views here, and I think it is a helpful part of the discourse. I do not believe, as he does, that AI is an inevitable doomsday device, but I do think his position is an important one to explore.
I plan to follow up with voices who might counter his particular perspective and legibly suggest the ways he might be wrong. Because I want him to be wrong.
As always, thanks for listening.
A small ask:
The irony is not lost on me in trying to critique the algorithms while still being dependent upon them to reach the right audience.
If you do enjoy the show, please consider becoming a paid subscriber, share it with a friend or drop us a rating.
You can follow us here:
* Apple Podcasts
* YouTube
* Spotify
* Instagram


Get full access to Into The Machine at tobias.substack.com/subscribe

Dr. Steven Pinker is a cognitive scientist at Harvard, a 2x Pulitzer Prize finalist, and author of notable works like The Blank Slate, How the Mind Works, The Better Angels of Our Nature, and Enlightenment Now.
Today we discuss his research on the psychology of common knowledge and what it means for democracy in a time of infinite AI slop and social media confusion.
We also dive into deeper philosophical questions on AI consciousness and rogue AI — note that this was recorded before the recent Hugging Face incident, making our commentary especially urgent.
His excellent most recent book: When Everyone Knows That Everyone Knows… inspired me to write The Noisy Room — an interactive essay that explains how we might fix the common-knowledge trap sprung by social media.
Programming note: We’ve been off working hard on a great new season of content. I want to offer enormous gratitude to our subscribers for their steady support. Thank you.
A small ask:
The irony is not lost on me in trying to critique the algorithms while still being dependent upon them to reach the right audience.
If you do enjoy the show, please share this episode with a friend and drop us a rating.
You can follow us here:
* Apple Podcasts
* YouTube
* Spotify
* Instagram
Thanks for listening, and please do subscribe.
-Tobias


Get full access to Into The Machine at tobias.substack.com/subscribe

What happens when the good people leave social media?
Chloé Valdary walked away from 110,000 followers on social media. Not a quiet, lurking exit — but a hard delete.Chloé was a force of nature on Twitter. She built a huge following helping people navigate some of the platform’s thorniest issues with grace and empathy. While many cancelled and called out, Chloé dove headfirst into the hardest conversations offering deep reflection and a dose of savvy humanizing. From George Floyd to Gaza, she was there, helping to deescalate.
Then she left.
In this episode, we unpack what it means when moderates like Chloé leave the platforms that dominate our public conversation.
Who is left? Usually it’s the most polarized and extreme users, who actively own the discourse after that point.
We talk a lot about social media as an Outrage Machine (and my book — of the same title), the hidden incentives shaping our behavior on these platforms, and the mental health cost of constant moral performance.
* If a social platform is a toxic place, what are the obligations we have to the public commons in staying there?
* Are big digital spaces like social media structurally incapable of supporting prosocial conversations?
* Can we protect our mental health while still participating in online political debate?
This conversation explores the trade-offs.
Chloé’s Substack: Musings with Chloé
Chloé’s work: Theory of Enchantment
A small ask:
The irony is not lost on me in trying to critique the algorithms while still being dependent upon them to reach the right audience.
If you do enjoy the show, please share this episode with a friend or drop us a rating
Into The Machine is a listener & reader supported project. If you enjoy these conversations, please consider becoming a paid subscriber. It allows for me to spend more time and effort researching new guests and making this a real thing.
You can follow us here:
* Apple Podcasts
* YouTube
* Spotify
* Instagram


Get full access to Into The Machine at tobias.substack.com/subscribe

Will people stop dating humans?
Over a decade ago, the movie Her portrayed a world where by where large numbers of people opted out of dating, and instead turned to AI for their emotional needs.
Justin McLeod is the founder and CEO of Hinge. He has spent more than a decade thinking about how algorithms shape dating, choice, and human connection.
In this conversation, we look at the rise of AI companionship, how business incentives affect these systems, and where AI-driven intimacy may be heading.
I press Justin on the risks of people turning to AI instead of human relationships, how much autonomy we give to algorithms, and what responsible design might look like.
This episode maps out the key questions we should be asking as we lean further into machine-mediated love.
I hope you enjoy.
A small ask:
The irony is not lost on me in trying to critique the algorithms while still being dependent upon them to reach the right audience.
If you do enjoy the show, please share this episode with a friend and rate us on apple podcasts.
You can follow us here:
* Apple Podcasts
* YouTube
* Spotify
* Instagram


Get full access to Into The Machine at tobias.substack.com/subscribe

Today I’m releasing a conversation with Tristan Harris.
Tristan is the founder of the Center for Humane Technology and one of the leading voices warning about how runaway AI might destabilize society. He starred in (and co-produced) the Netflix documentary The Social Dilemma.
I’ve known Tristan for a long time, and this is one of the best conversations we’ve ever had—public or private. I press him on major risk scenarios, what we can expect AI labs and legislators to do in the face of AGI, and what he thinks can actually be done right now to ensure these systems stay maximally beneficial to humanity.
I left this conversation a bit more hopeful understanding what kinds of solutions are available. I hope you do too.
In this episode we talk about:
* Creepy new AI capabilities: new models using unwitting humans to send encoded messages to other AIs.
* How it might all go down: a real-world near-term disaster scenario with runaway self-replicating AIs.
* How to bypass race dynamics with China and other powers accelerating AI capabilities.
* Designing systems for wisdom: alternative paths for designing and training Socratic AIs.
A small ask:
The irony is not lost on me in trying to critique the algorithms while still being dependent upon them to reach the right audience.
If you do enjoy the show, please share this episode with a friend and drop us a rating.
You can follow us here:
* Apple Podcasts
* YouTube
* Spotify
* Instagram
Thanks for listening, and please do subscribe.
-Tobias
Into The Machine is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.



Get full access to Into The Machine at tobias.substack.com/subscribe