How do you build a system for turning wild ideas into world-changing innovations? Astro Teller, Captain of Moonshots at X, The Moonshot Factory, has spent over 15 years leading Google’s audacious innovation lab—the birthplace of Waymo, Google Brain, and other breakthrough projects.

In this special episode, recorded live in Austin at last year's SXSW, Astro shares the playbook to create a moonshot factory.

(I'm at this year's SXSW right now and you'll hear all about it soon. If you are here, drop me a line and let's meet up!)

What You’ll Learn in This Episode:

• The “Train the Monkey First” approach to innovation
• Why audacity, humility, and intellectual honesty are key to moonshots
• How your org can get more 10x (not +10%) outcomes — and how to avoid the “innovator’s dilemma”
• Why you should “greenlight everything” and then redlight most projects quickly, following kill criteria you’ve agreed to in advance
• Where X is placing bets today, including climate-tech, modernizing the electric grid and bioengineering

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

Greetings from SXSW, where I'm learning, recording, and eating... You'll hear all about it soon... For now, enjoy this short, sweet, and geeky bonus episode.

Have you seen that weird graph about all the jobs that AI is going to kill? It looks like an ink blot or a Rorschach test... It's from an Anthropic report and it's really making the rounds. If you follow tech stuff on social media you've probably seen it. The report is interested, but I'm convinced people are only sharing it because the graph looks cool and people will think they're smart if they share this inscrutable data visualization...

Anyway, here's a very short excerpt of my upcoming interview with Paul Ford (@ftrain), one of my favorite tech writers and the founder of Aboard. He and I took a break from talking AI and such to geek out on this data visualization and why it's so bad, plus I told him about how I used AI to make my own version of a radar graph (about how many, and which kinds of, tacos I will and could theoretically eat in Austin).

---
Subscribe to the Future Around & Find Out newsletter!

So why is one of the world’s leading AI researchers teaching AI to understand pain and suffering? Well, Daniel Hulme says that if we build an empathetic AI, perhaps even a conscious one, then we’ll be safer. His hypothesis is that a "zombie" AI will eat our brains, but an empathetic AI would stay aligned with us. So he's building this "antivirus" (with AI, of course) and he's very aware that this sounds crazy or like "something from Marvel."

That's just some of what broke my brain in this conversation with one of the world's top AI researchers and founders. And Daniel has serious credibility, so I'm not dismissing the threat he sees — you know, the one where we all get turned into paperclips.

Daniel sold his company Satalia to WPP, where he now serves as Chief AI Officer. He’s just founded Conscium, which verifies that AI agents are safe and can do what they promise — and is also researching consciousness and pain. Some of the world’s leading AI thinkers are on the advisory board and Daniel has been in this space for decades: we’ll talk about why, for his PhD, he studied bumblebee brains (yes, really — and it's deeply relevant).

We get into:

• His unified theory of consciousness — his "color wheel" model — and why he thinks consciousness only exists in motion
• Why he believes large language models are ultimately a dead end — and what neuromorphic computing could replace them with
• What bumblebee brains can teach us about building AI that's up to a thousand times more energy efficient
• Why he calls today's AI agents "intoxicated graduates" — and says companies should spend 80% of their time testing them
• The concept of "mind crime" — the idea that we could build conscious AI and accidentally put it through horrendous suffering without realizing it
• His vision of a "protopia" — where AI makes food, healthcare, education, and energy so abundant that people are freed from economic constraints to pursue what actually matters

We future around and find out a lot in this one!

---
Chapters

• (01:39) - "Would a conscious superintelligence be safer than a zombie one?"
• (03:37) - The paperclip problem is not hypothetical
• (05:06) - Conscium's mission — AI safety for humans and for AI themselves
• (08:50) - "I think I've got my head around consciousness"
• (11:57) - The color wheel model — why consciousness only exists in motion
• (13:58) - Teaching AI morals through evolution, not guardrails
• (17:23) - "Hey Claude, are you conscious?" — how do you test for that?
• (21:07) - What bumblebee brains can teach us about building better AI
• (24:14) - "I think we are completely scaling wrong"
• (29:43) - Why Daniel calls AI agents "intoxicated graduates"
• (32:48) - Companies should spend 80% of their time testing agents
• (38:19) - "What would you do if you were economically free?"

• Conscium
• Daniel Hulme on Wikipedia
• Daniel on LinkedIn

---

So how do Kwaku's kids know that it's FAFO Friday? "They're like, 'oh, we know you're doing the podcast 'cause we just hear you cackling through the walls.'"

So laugh along with Kwaku and me today as we work our way through a quick victory lap (stuff we said would happen last week happened!), why Sam is like that desperate guy at the bar who refuses to go home alone, quantum computing explained via children's literature, why the Jetsons are not reason enough for us to build humanoid robots, robot choreography (are we human or are we dancers?), wen self-driving cars in NY?, riding a wave of green lights up Manhattan's third avenue at 2 AM, artificial wombs and other moonshot off-shoots, and the real origin of Velcro (AI lied to me about it).

Plus... goat ranches, breakfast tacos, and what we're most excited about heading into SXSW. It's a choose your own adventure kind of day.

Chapters

• (01:24) - Victory Lap — We Called It
• (03:35) - OpenAI's Bar Guy Energy
• (06:38) - Waymo, Robot Choreography, and Green Light Waves
• (10:16) - Self-Driving Cars vs. New York Politicians
• (13:13) - What We're Most Excited About at SXSW
• (15:41) - Quantum Computing: Choose Your Own Adventure Edition
• (18:01) - Dire Wolves, Moonshots, and Tech Nobody Sees Coming
• (24:07) - Why Do Robots Need to Look Like Us?
• (29:22) - The SXSW Way-Back Machine
• (36:08) - Increased Regulation: Past, Present, or Future?

• Follow Dan on LinkedIn
• Get the free Future Around & Find Out newsletter
• Become a paid subscriber and help future proof the podcast!

Sponsor the show?

• Are you looking to reach an audience of senior technologists and decision-makers? Email me: dan@modernproductminds.com

---
Music by Jonathan Zalben

So, there are dire wolves living on Earth again. They were “de-extincted” by Colossal Biosciences. And today on the show their Chief Science Officer joins me to share her view on why the de-extinction matters — not as a science project, but because it will help solve problems that threaten every species on earth, including us.

Beth Shapiro is the Chief Science Officer at Colossal Biosciences, and she helped to bring back the dire wolf or, as others call it, a gray wolf with 20 genetic edits. There is a fierce debate about what de-extinction even means, and we discuss that, but whatever you call them, there are now three big wolves living in an undisclosed location and they wouldn't be there if not for the DNA that Beth and her team edited. Colossal is also working to bring back the wooly mammoth, the Tasmanian tiger, the dodo and other animals that have long been extinct. Why?

Listen to find out…

Chapters:

• (01:19) - The Most Oprah Question Beth's Ever Been Asked
• (03:04) - Moonshots Require You to Create a Giant List of Problems
• (04:19) - The Things We’ll Solve Along the Way, a la the Original Moonshot… to the Moon
• (05:57) - Beth’s Journey: From Broadcast Journalism to Ancient DNA
• (09:13) - How a Sediment Core Solved a Mammoth Mystery
• (11:36) - Why Charismatic Animals Matter (a.k.a. Why Riz Is Everything)
• (12:38) - What’s Up With Romulus, Remus, and Khaleesi?
• (14:19) - But Are They Really “Dire Wolves”? The Controversy Over 20 Genetic Edits
• (21:45) - Should We Do This? Beth's Ethics Framework for Builders
• (23:51) - Advice for Moonshot Builders
• (25:10) - Why We Want Dodos, Mammoths, and Thylacines Back

• Colossal Biosciences
• Beth Shapiro
• PopTech -- a conference I love!

Support Future Around & Find Out

• Follow Dan on LinkedIn
• Get the free Future Around & Find Out newsletter
• Become a paid subscriber and help future proof the podcast!

Sponsor the show?

• Are you looking to reach an audience of senior technologists and decision-makers? Email me: dan@modernproductminds.com

---
Music by Jonathan Zalben

Murderbots, mass layoffs, and media takeovers — all in one news cycle. Anthropic told the Pentagon "we will not accede." Block cut half its workforce overnight. And the Paramount-Warner Brothers deal raises real questions about who's running the media now.

Also, thanks to Nicolás Maduro's fashion sense, Dan's 13-year-old is being called Lil Tator at school and honestly? The kids are all right.

Happy FAFO Friday!

Here's some of what Kwaku Aning and I get into:

• (00:00) - Three Stories Broke Last Night
• (03:16) - Anthropic Tells the Pentagon No
• (06:24) - Murder Bots, But Human in the Loop
• (07:00) - The Pentagon's Friday Deadline
• (09:28) - Why This Is a Huge Win for Anthropic
• (10:50) - The War for AI Talent
• (12:57) - Is the Administration Losing Steam?
• (15:05) - The Paramount-Warner Brothers Deal
• (17:36) - Who Controls the Media Now?
• (21:13) - CNN, Independent Media, and the Employee Perspective
• (23:55) - Block Lays Off 4,000 People
• (24:14) - The Citrini Research Fiction That Tanked Stocks
• (27:49) - AI Washing and the Real Reason for Layoffs
• (30:11) - Will Vibe Coding Replace Real Companies?
• (33:27) - Mid-Roll Break
• (34:41) - Past, Present, Future: State-Controlled AI
• (35:18) - Past, Present, Future: Independent Media
• (38:03) - — SLAPP Lawsuits and Creator Protections
• (40:23) - — Past, Present, Future: Knicks Championship
• (41:44) - — Come See Us at South by Southwest!

Of all the industries AI will transform, Kira Radinsky believes chemistry and biology will change the most.

Kira is the co-founder and CTO of Diagnostic Robotics, which uses AI to automate the administrative work that's crushing healthcare teams — so clinicians can actually focus on patients. She's also the co-founder of Mana.bio, where they're accelerating drug discovery by orders of magnitude.

She'll tell you she's terrible in the lab. Not because she isn't brilliant, but because she can't pipette without killing the cells. So she’s thrilled that thanks to her skills in data and AI she was able to realize her childhood dream of being a scientist:

But this episode is about more than healthcare. It's about how to build systems that get smarter over time — feedback loops, causal inference, incentivizing algorithms to take risks, and knowing when to optimize for ROI instead of accuracy. Lessons that apply whether you're building in biotech or not.

We cover:

• How growing up Jewish in Soviet Ukraine — and fleeing to Israel just before the Gulf War — shaped Kira's obsession with predicting the future
• How she built a system that successfully predicted real-world events, including Cuba's first cholera outbreak in Cuba in 130 years
• How Mana.bio is using AI to build "rocketships" that deliver drugs to the right cells — and how they've done in three months what used to take 20 years
• Why predictions are only valuable if there's something you can do about them — and why that makes healthcare an ideal field for AI
• How to incentivize algorithms to make bolder predictions (it's easy to predict there won't be an earthquake today; it's much harder to say there will be)
• Why causal inference is the most underrated tool in machine learning right now
• How healthcare AI can perpetuate racial bias — and what builders need to do differently

Note: this interview originally aired in October 2024.

Chapters:

• (01:44) - Why predictions are so important to Kira: lessons from fleeing Soviet-era Kyiv
• (05:10) - Building a prediction engine from 150 years of news
• (08:35) - How Kira predicted the Cuba cholera outbreak
• (09:50) - Returning to biology by way of data
• (12:50) - Predicting healthcare outcomes by finding your patient's twin
• (17:53) - The racial bias hiding in healthcare AI
• (19:15) - Building Mana.bio and accelerating drug discovery
• (24:33) - "In three months, what did what used to take 20 years"
• (31:44) - Builder tips: ROI, causal inference, and teaching algorithms to explore
• (35:07) - Planning: Where generative AI needs improve

• Kira Radinsky on LinkedIn
• Diagnostic Robotics
• Mana.bio

Support Future Around & Find Out

• Get the free newsletter
• And consider becoming a paid subscriber and help future proof this thing!

Sponsor the show?

• Are you looking to reach an audience of senior technologists and decision-makers? Email me: dan@modernproductminds.com

---
Music by Jonathan Zalben

You probably know by now that AI is the definition of mediocre. As in: it’s the average of everything it’s been trained on. So how do you get beyond average? How do you build a moat?

It certainly doesn’t seem to be via the models. While there are models of the month (hey, Opus 4.6, my new friend!), they seem to be pretty swappable.

So, the model ain’t it. But proprietary data (e.g. an AI that knows you really well), yes! Or doing something really hard in the real world (think: Waymo self-driving cars). Maybe via trust and safety (Anthropic is certainly making a play here). Or... how about via amazing design and good taste.

Remember when ChatGPT first came out and everyone derided “AI wrappers”… well, maybe a wrapper isn’t so bad, assuming you can differentiate on one or more of the above.

Luke Des Cotes is the CEO of MetaLab, the agency famous for designing interfaces, including early versions of Slack and Coinbase, so don’t be shocked when you hear him say that great design can be your moat.

MetaLab is working with a host of AI companies (another shocker), including Windsurf (AI + code), Suno (AI + music), Pika (AI + video), and more…, which is why Luke's take on AI surprised me. He's not rah rah. He's pretty judicious actually. Luke has questions about AI's costs and appropriateness for lots of use cases like those involving kids, but mostly he objects to its mediocrity.

On this episode we discuss what it takes to go beyond.

We also get into:

• Why vibe-coded software isn't changing the world anytime soon
• Why Shopify acquired a design agency right after telling employees to justify their existence against AI
• How MetaLab designers are using AI to prototype in hours instead of weeks
• The talent market for zero-to-one designers — and why they're harder to find than ever
• Landlines, brick phones, and how parents are fighting back against always-on kids

Chapters

• (01:10) - "It's a race to the mean"
• (03:10) - "How do you create emotional resonance?"
• (05:33) - AI companies are burning money
• (08:44) - Speed to good enough
• (13:51) - Is the chat here to stay or a temporary fad?
• (17:43) - It’s hard to find great 0 to 1 design talent
• (22:28) - Seemingly conscious AI
• (25:05) - Kids, landlines, and fighting always-on culture
• (27:21) - Sounds like science fiction, but is here now…

Links & Resources

• Luke Des Cotes on LinkedIn
• MetaLab

Support Future Around & Find Out

• Get the free newsletter
• And consider becoming a paid subscriber and help future proof this thing!

Sponsor the show?

• Are you looking to reach an audience of senior technologists and decision-makers? Email me: dan@modernproductminds.com