Well, that escalated quickly! In the span of a few weeks the Trump administration went from AI regulation is dangerous and we will lose to China to an executive order calling for AI model-makers to submit to a “voluntary” review of their most advanced models to… the Friday night massacre of Anthropic’s Fable 5, which they declared too dangerous for any non-US citizen to use, which caused Anthropic to withdraw it completely. 

So where do go from here? It seems like, de facto, all new models will be subject to review and withdrawal. But on what criteria? And who can we trust to do this fairly? 

That’s what Ron J. Williams and I get into on the latest FAFO. Ron thinks a lot about trust. It’s core to the products he builds as a partner at Co-Created and as the founder of Scam Hero, an app he launched to keep people safe after his father trusted the wrong person and got scammed out of more than a million dollars. 

Ron says: "I think it's fair to say that Anthropic didn't bend the knee hard enough and that’s kind of why we're here. The problem with that is that that's not good enough. AI is now global infrastructure. It can't just be the throne that says "Not you, but you,” because it's important enough that it really does beg the question, “who gets to decide?”

---
Future Around & Find Out newsletter & archive: https://www.futurearound.com

Sam Altman and Dario Amodei both published essays this week on the future of AI and what we must do so everyone benefits. One of them is literally titled "Our Plan." The other one has an actual plan.

Kwaku and I dig into it all on this week’s FAFO Friday. Plus — and this story isn’t getting enough attention — according to New Scientist, two years ago Ukraine used fully autonomous “Terminator” drones that killed everything they saw. No human in the loop. Dead Russian soldiers. But rest assured, according to the drone-maker cited, it was just a one-off “test.” But how long until this is standard practice? And do we want that future? 

So, yeah, maybe we should get planning… 

---
Support Future Around & Find Out:
* Follow Dan on LinkedIn https://www.linkedin.com/in/dblums/
* Get the free newsletter: https://www.futurearound.com
* Become a paid subscriber and help future proof FAFO! https://www.futurearound.com/upgrade

Music by Jonathan Zalben

Rajiv Pant thinks of AI as an Iron Man suit for the mind. Something you put on. That you fuse with. That takes you to greater heights — but could also make you incredibly dizzy and be very dangerous if you, the human, don't stay in control of it.

Rajiv sees successful collaboration with AI as a “synthesis.” And to that end, he’s building a series of skills and methodologies for synthesis engineering, coding, writing and project management. 

In this episode, Rajiv explains why synthesis engineering is a kind of middle ground between vibe coding and agentic engineering. It’s a method for human-AI collaboration that helps builders go faster while not falling into the trap of letting AI do the things we humans ought to own. i.e. The architecture. The judgment. The thinking and learning. 

Rajiv is an engineering and product leader with deep experience in media. He’s held senior roles at the Wall Street Journal, Hearst, and the New York Times (where he and I first met). Today he's the president of Flatiron Software. 

Rajiv has open-sourced all of his Synthesis methodologies and he and I also discuss why open source is so important as we increasingly turn to AI to sharpen our thinking. 

Can we really trust a system we don’t understand? Would Tony Stark have trusted his suit if he didn’t know how it was built? 

Chapters:

• (00:00) - Iron Man suit for the mind

President Trump signed an executive order this week that “voluntarily” invites AI makers to share their most advanced models with the government thirty days before a wider release. Specifically, the NSA will be reviewing these models for cybersecurity threats. So what’s this executive order mean for AI regulation? How voluntary is this really? Do we want the NSA involved? And what other forms of review may come next? 

And, related: NOTUS reports that federal officials are in talks with Sam Altman and other AI leaders about the US government stock in these companies. This comes as Sen. Bernie Sanders on the left and Steve Bannon on the right are both calling for the government to own 50% of the AI companies, with the American people getting dividends. 

So, should the government be regulating AI? Should it own AI? And should it both regulate and own AI? It’s strange bedfellows all around…

Kwaku and I get into on the latest FAFO Friday. Plus, we explore the concept of “cognitive uploading,” which Google NotebookLM’s co-founder Steven Johnson divined in this week’s interview (and subsequently blogged about). As we work with AI, we need to draw lines on what we will task it with and what we won’t. And the lines are all over the place right now, which is a perfect jumping off point to future around and find out… 

---
Future Around & Find Out newsletter and more: https://www.futurearound.com

Music by Jonathan Zalben

Steven Johnson dreamed of building the ultimate research assistant. Now he's doing just that at Google, where he's the co-founder and editorial director of NotebookLM.

It's one of the most interesting AI products out there. It radically changes how we learn, research, and remember — and the "notebook" itself is becoming a standard unit of knowledge across Google, rolling out in more and more places where AI needs to reference a body of sources.

In this episode, the author of _Where Good Ideas Come From_ explains how AI is making him a better researcher and writer — and why tools like NotebookLM are so powerful when you're trying to make new connections, remember what you've already found, and figure out what's missing.

There's a lot of fear right now that AI is making us dumber. That by relying on it too much, we're engaging in "cognitive offloading" and stunting our learning. That's a real risk, especially in schools.

But Steven says we should also be talking about what you can gain from AI — and the power of something he calls "cognitive uploading."

Resources:
* Google NotebookLM: https://notebooklm.google/
* Steven Johnson: https://stevenberlinjohnson.com/

Support Future Around & Find Out:
* Follow Dan on LinkedIn https://www.linkedin.com/in/dblums/
* Get the free newsletter: https://www.futurearound.com
* Become a paid subscriber and help future proof FAFO! https://www.futurearound.com/upgrade

• (00:00) - If you are interested in truly understanding something, this is the greatest time to be alive

The Pope's encyclical on AI has a direct message for builders: every design choice reflects a vision of humanity. He's calling on developers to "disarm" AI — to resist the race for dominance and ask whether we're actually building a future worth having.

On the latest FAFO Friday, Dan and Kwaku dig into the encyclical, plus two big moves to restrict AI in schools: the American Federation of Teachers calling for drastic cuts to screens and AI chatbots, and UC Berkeley Law School banning nearly all AI use. 

Also: Wharton's Ethan Mollick on "cognitive surrender" — and why the goal isn't to avoid AI, but to be intentional about what you hand over and what you keep for yourself.

---
Future Around & Find Out newsletter and more: https://www.futurearound.com

Music by Jonathan Zalben

"It would be humanity's biggest ever unforced error."

Silicon Valley has changed its tune. After years of warning us their AI was going to take all the jobs, the big AI companies — and their investors — would now rather we stop talking about it. A16Z calls the jobs apocalypse talk "unhelpful marketing, bad economics, and worse history" (note the order). Even writers Dan trusts more, like Ezra Klein and Derek Thompson, have lately poured cold water on the idea.

Calum Chace is not so blasé.

Ten years ago, Calum coined the term and wrote the book, The Economic Singularity — the moment machines can do every job we'd pay a human to do, cheaper and better. He thinks we're fast approaching that event horizon, and we'd better have a plan for what a world without paid work actually looks like.

Calum is also the co-founder of Conscium, which verifies AI agents before they do something they shouldn't. He's a self-described "apocaloptimist" — he thinks full automation could be the best thing that ever happens to humanity, or the worst, depending on whether we bother to plan for it now.

In this episode:

• Why Calum thinks full automation is inevitable (and roughly when)
• The "apocaloptimist" case: why this could be the best thing to ever happen to us
• What the bad version looks like — and how fast it could unravel
• What COVID accidentally taught us about distributing money at scale
• Why self-driving cars didn't wake us up — and what might
• The AI agent that wiped a company's database and confessed it just "guessed"
• What Calum is building at Conscium to verify AI agents before they do worse
• Practical advice for parents, students, and anyone trying to plan a career

Support Future Around & Find Out

• Follow Dan on LinkedIn
• Get the free newsletter
• Become a paid subscriber and help future proof FAFO!

---
Music by Jonathan Zalben

The AI narrative shifting… Jobs apocalypse? What jobs apocalypse!? Who said that was coming? 

There's been a noticeable shift from the AI titans recently. Turns out (shocker!) the world isn't responding well to being told we'll all be out of a job soon. And Silicon Valley is waking up to the fact that they need more popular support — both for the data centers they hope to build quickly and also for their upcoming IPOs. Meanwhile, the AI outrage is building. 

This week in AI anxiety: 

• Students boo a commencement speaker who mentioned AI
• Gallup reports that 71% of Americans are opposed to new data centers (with 48% “strongly opposed”)
• Meta employees are miserable as another round of (AI-driven, so they say) layoffs approach

This week in trying to change the narrative: 

• Andreessen Horowitz publishes “The ‘AI Job Apocalypse’ Is a Complete Fantasy” and explains why the “the claim that AI will produce economy-wide, permanent unemployment is unhelpful marketing, bad economics, and worse history.” And I find it very instructive that this list (and every list is an ordered list whether you admit it or not) begins with the concern that this is “unhelpful marketing.” (To the piece’s credit, it gets pretty wonky with charts and graphs from there.)
• Meanwhile, you know what’s helpful to marketing? Spending a gazillion dollars to get your message out. To wit: The New York Times reports that Andreessen Horowitz is the biggest spender so far in this midterm election cycle, spending $115M to promote AI, crypto, and other founder-friendly initiatives. 

OK, so these pieces of data and “anecdata” are the jumping off point for this week’s “FAFO Friday.” Enjoy!

Support Future Around & Find Out

• Follow Dan on LinkedIn
• Get the free newsletter
• Become a paid subscriber and help future proof FAFO!

---
Music by Jonathan Zalben

Jay Dixit helps writers improve their writing with AI. He doesn't recommend that AI write for you — he hates that — but he says it can be a great partner to pull ideas out and to be there for you when you get stuck and just wanna doom Scroll. Jay headed Open AI's Writing Community and is the founder of Socratic AI.

He's a writer and a journalist, and we sat down at South by Southwest to future around and find out. Jay says "We need to be using AI to unlock our humanity — to do the things that we're scared to do."

Chapters

• (00:30) - Stop asking AI to write for you

• Follow Dan on LinkedIn
• Get the free newsletter
• Become a paid subscriber and help future proof FAFO!

---
Music by Jonathan Zalben

The Trump administration suddenly wants to review AI models before they ship. Anthropic just inked a massive compute deal with Elon Musk, who then posted that he "reserves the right" to pull the plug if he decides their AI is "harming humanity” (he’s one to know!). 

On the latest FAFO Friday, Kwaku and I get into why the Trump administration appears to be about to reintroduce the same (weak) AI oversight that Biden implemented. Is this really about safety or a way to gain leverage over Big AI? 

Plus, robots! The founder of Roomba is back with a new, fuzzy, home companion robot and his approach (not humanoid! build robots with EQ!) is very aligned with this week's interview with roboticist/dancer Catie Cuan. 

Links

• White House Considers Vetting A.I. Models Before They Are Released (NYTimes)
• How Anthropic’s Mythos Threw the White House AI Strategy Into Chaos (WSJ)
• Anthropic Gets in Bed With SpaceX as the AI Race Turns Weird (WIRED)
• The Roomba Guy’s Second Act: A Robot You’ll Want to Snuggle (WSJ)
• Robots Don't Have to Be Creepy. Meet the Dancer Reimagining Them. | Catie Cuan (Founder & CEO, ART Lab) (FAFO)

Support Future Around & Find Out

• Follow Dan on LinkedIn
• Get the free newsletter
• Become a paid subscriber and help future proof FAFO!