Episode #45: Three Breakthroughs: 1 billion robots by 2030, Sodium Batteries?, 529s are dead

Frequently Asked Questions

FAQ

• Where is the transcript for this webinar?

  Samantha Herrick:
  We’ve got a cool one for you today—AI and Khan Academy, salty batteries, and your own Rosie robot.

  Mike Collins:
  But obviously here at AV, we are also big believers that where companies go to die, which is to innovate in the area of education, that there is new potential.

  Michael Peri:
  If sodium-based batteries can truly be scaled up and brought to market, it’s going to dramatically reimagine how we operate with this stuff today.

  Mike Collins:
  These things tend to be introduced, hyped, disappointed, and then become part of life.

  Samantha Herrick:
  Hi everyone. Welcome back to this episode of the Tech Optimist. We’ve got a Three Breakthroughs episode planned for you today. There are some really awesome topics that come up in this conversation—just really innovative topics and insightful discussion.

  So of course, we’re going to have Mike Collins here on one end of the table—founder and CEO of Alumni Ventures—who will guide us through. His partner in crime is partner Michael Peri, but we also call him Mike. And then, of course, you’ll hear my voice—I’m Sam, the guide and editor for the Tech Optimist podcast.

  A little bit more about today’s conversation: one breakthrough they talk about is AI and education. There’s a really cool conversation between Bill Gates and Sal Khan, CEO of Khan Academy, discussing how AI is changing education.

  Then Mike Peri brings up a breakthrough involving sodium-based batteries, which in the long term are supposed to be more efficient, cost-effective, and better for the environment overall.

  And lastly, we wrap up with humanoid robots. The discussion includes Tesla’s robot, a robot from Figure, and a few other snippets sprinkled throughout.

  I won’t give away too much more, so without further ado, here’s Mike and Mike as they share the breakthroughs that caught their attention this week. We hope you enjoy it.

  As a reminder, the Tech Optimist podcast is for informational purposes only. It’s not personalized advice and it’s not an offer to buy or sell securities. For additional important details, please see the text description accompanying this episode.

  Mike Collins:
  Tech Optimist podcast, brought to you by Alumni Ventures. On this show, I get together with one of our big thinkers—today it’s Mike Peri—and we talk about breakthrough technologies and innovations with the potential to put a little dent in the universe. I think I’m going first today.

  Michael Peri:
  All right.

  Mike Collins:
  I just finished Sal Khan’s new book—it came out a month or so ago—called Brave New Words. Sal got an early peek into ChatGPT and has been an incredible entrepreneur, really flipping the script on education. I come from a long line of teachers, and his work has truly made a difference.

  He speaks eloquently about something I think many of us are interested in: education, and specifically how AI could be a game changer—and how he believes it will be a game changer.

  I encourage people to check out the book. He also does a really good interview you can find online with Bill Gates, where they talk about it. The Gates family has been early and consistent supporters of Khan Academy.

  Here at AV, we’re big believers that education has historically been an area where companies go to die when they try to innovate—but there’s new potential now. I think I’m going to title this segment: “Stop Investing in Your Grandchild’s 528.”

  Michael Peri:
  529.

  Mike Collins:
  529, excuse me. Definitely don’t invest in a 528! But your 529—which, by the way, I personally did for my three kids, starting early and consistently—was a great financial move for my family.

  But I believe if you’re giving birth this year and looking 17 years out, I can make a pretty strong prediction: your child’s educational experience is going to be very different in 15 years.

  Samantha Herrick:
  Okay, so along these lines, I did a little digging and found the interview between Bill Gates and Sal Khan, CEO of Khan Academy. It’s a fascinating clip from Bill Gates’s podcast Unconfuse Me, where they talk about how OpenAI approached Khan Academy to help train the AI assistant model to be able to do the AP Biology exam.

  When I was in high school taking AP classes, I can’t imagine how many doors would’ve opened if I’d had AI as a tool to help me study and prepare for exams.

  This conversation is really interesting. I’ve cut it down—it’s not the full eight minutes—but here are some snippets. If you want the full video, we’ll leave the link in the description and show notes.

  First, let’s take a quick break for an ad, then we’ll jump into the conversation with Bill Gates and Sal Khan. After that, we’ll return to Mike and Mike discussing how education is changing and how it affects them personally as parents.

  Matt Caspari:
  Hey everyone, just taking a quick break to tell you about the Deep Tech Fund from Alumni Ventures. AV is one of the only VC firms focused on making venture capital accessible to individual investors like you. In fact, AV is one of the most active and best-performing VCs in the US, and we co-invest alongside renowned lead investors.

  With our Deep Tech Fund, you’ll have the opportunity to invest in innovative solutions tackling major technical and scientific challenges—companies with the potential to redefine industries, create a more sustainable future, and deliver significant returns.

  If you’re interested, visit us at av.vc/funds/deeptech. Now back to the show.

  Bill Gates:
  Welcome to Unconfuse Me. I’m Bill Gates. Let’s talk about AI. You and I have both been lucky enough over the last six months to engage with both Microsoft and OpenAI and get early access.

  I remember some of the best examples of getting AI to do fun things were when you came up to see me and we brainstormed about this. You were the one who said, “Hey, if you tell it to write a speech like various politicians—including Trump and others—it’s stunning how it captures their voices.”

  Bill Gates:
  So give us how you first were using the AI, and then this is super timely because you’ve just recently come out with Khanmigo—if I’m saying that right.

  Sal Khan:
  That’s right, very good. Yeah, I mean, the OpenAI folks reached out and they said, “Hey, we’re a couple of weeks away from having the first training of our model,” and they wanted to reach out to Khan Academy for two reasons. One was they said, “We want to make this really good at AP biology.” And I only found out later—I don’t know if this is true—they told me you gave them the challenge.

  Bill Gates:
  That’s right. In June, they kept showing me this thing and I was like, “Yeah, it’s kind of an idiot savant. I don’t think it’s practical. Why don’t you see if it can do the AP biology exam, and I’m not going to pay any attention until you can get a five.” And I thought, okay, that’ll give me three years to work on HIV and malaria.

  And then it was so bizarre because Sam Altman and Greg Brockman, in like August, said, “Hey, we want to come show you this thing.” And so it was early September when there were like 30 people at my house, and I said it was the most stunning demo I’ve ever seen in my life. I mean, right up there with seeing the Xerox PARC graphics user interface that set the agenda for Microsoft for about 15 years. This demo was so surprising to me—the emergent depth—that as they scaled up the training set, its fluency.

  And you have to say, understanding that computers could not read in the sense humans do, and it couldn’t write in the sense humans do, and now—with lots of footnotes about hallucination and things like that—but I’m still personally in a state of shock at, wow, it is so good. And therefore, let’s see where we can put it to good use.

  Sal Khan:
  Yeah, no, 100 percent. And so thank you for that challenge to them. I think they came to us because we have a large library of AP biology questions, etc. They’re like, “Hey, can we use that to either evaluate or train?” And at the time, I was like, “Well, what’s in it for Khan Academy?” And they’re like, “Well, maybe you get access to the model.”

  And I too was skeptical. I had seen GPT-3 at that point, and it was cool, but I didn’t see how we were going to apply it. Two weeks later, they showed us the AP bio question. They said, “So Sal, what’s the answer to this?” And I was like, “Okay, I think it’s C.” And it said, “Oh, the answer is C.” And I was like, oh, that’s interesting. I started getting a few goosebumps and I said, “Well, ask why that’s the answer.” It explained it.

  Bill Gates:
  Oh yeah. It’s so good.

  Sal Khan:
  Oh yeah. And I mean, I think folks need to realize—because everyone had that moment with ChatGPT—but it was like that, but more, because GPT-4 is even better.

  Bill Gates:
  Way better, yeah.

  Sal Khan:
  And I said, “Can you say why the other answers aren’t correct?” It did. Then I said, oh. Then I started, I was almost shaking. Then I said, “Can you write 10 more questions like this?” Bam, bam, bam. And the first 10 I saw were all pretty good. I’m like, yeah, yep, that’s all legit.

  And then the implications for Khan Academy started to go through my mind. And then we did start to get into some of its hallucinations and some of its math errors in those early days. But that weekend, they gave me and our chief learning officer access to Slack. And we just—I couldn’t sleep—I was just having these in-the-rabbit-hole conversations with it.

  And then we had a hackathon for our team. We got about 40 people on our team under NDA, and we said, just come up with stuff. We were having the debates that everyone was having around, well, the information is not 100 percent airtight. The math isn’t great right now. The costs are not trivial. It can introduce bias. What’s the safety? How could you use people’s information, etc.

  But then we were starting to get it to work well as something that just helps you answer questions while you’re watching a video, to work well as a tutor. And then honestly, every 10 minutes we thought about it like, “Wait, it could also do this. It can also do that. It can also do that.”

  And so we said, well, what if we could make it so you could talk to any historical character? What if you can make it so it gets into a debate with you? What if you can make it so it doesn’t write your paper but writes it with you? What if it could do lesson plans for teachers? It could be the end of static curricula. I mean, just the imagination kept going.

  And by December, January, we had our team in full rapid prototyping mode. Just recently, we launched Khanmigo. And so far, we’re starting to titrate it out to the world, giving people access to it. But the feedback is very promising. What we’re hearing overwhelmingly from social media and the press is that they’re really happy that we’ve engaged in this and that we’re taking a safe approach where parents and teachers can monitor it. We have a moderation filter.

  OpenAI has also gone through great pains to make sure that things stay appropriate. You and I have talked extensively about the math issue, and we’ve done some things that I think make it quite robust on top of the things that the model does. And the costs are coming down dramatically.

  Michael Peri:
  This is super timely. I mean, Mike, I think you know this about us—we have a soon-to-be 4-year-old and a soon-to-be 1-year-old. And my wife and I have constant discussions around what the future of education will look like for them. From the costs associated with it—and yes, we started the 529 early too, so obviously cost is front and center—but even just how curriculum and learning are going to evolve over time.

  I think of the framework that OpenAI had put out. Remember there are like five stages of artificial intelligence, and I think we’re in the early innings. But as you think about levels 3, 4, 5—meaning systems that can take actions, AI that can actually aid in invention, and then AI that can do the work of organizations—how is that going to impact education?

  And our view is you have to have the human in the loop. So much of that is the fabric of our DNA and culture, the personal skills. So certainly timely—in the Peri household, we’re talking about this almost on a daily basis.

  Mike Collins:
  No, and I think a lot of families are. And the money, obviously—the unrelenting growth of the cost of education, from $20,000 kindergartens to million-dollar college education—is, one, unsustainable, and two, creates huge societal problems. And again, Tech Optimist—we believe that technology can be part of a solution.

  And I also believe that there is a lot more to education than just imbuing information. And the whole high-tech, high-touch thing, I think, is absolutely at work. But again, I think Sal talks about it very eloquently in a very balanced, real-world way. And I think he believes that we’re looking at really a fundamental change.

  And I think one way to think about it is everybody is going to have a personal university, customized to their particular interests, needs, learning proclivities. And it doesn’t take a huge leap—if you are using ChatGPT-4 every day like I do for my job—that it’s not a hard leap to think people are going to have a customized, personal AI partner in their educational journey from very early on, really throughout their whole life.

  Mike Collins:
  I mean, I just got back from Europe—a week vacation—and I nerd out on these places. And it’s like I had a personal translator, tour guide, history teacher, culinary expert just literally in my pocket. And so again, just to plug a few of our 15 portfolio companies in this space—because other partners will give me huge shit if I don’t—we are long on this, from professional education to doing research.

  I’m just going to have Sam save my breath. I’m going to put up a list of 10, just to pick a few of our portfolio companies that are coming at this from different angles.

  Samantha Herrick:
  Ask and you shall receive. Don’t worry, Mike, I got you. So some companies or some PortCos that are under AV that Mike was talking about—just a few examples—are Synthesis School.

  Synthesis School is an online program created at SpaceX that is now available to kids to encourage creative problem-solving and teamwork. From their website:

  “Education is too important to leave school. Using advanced teaching techniques and progress in AI, parents no longer have to. Synthesis Tutor does what school was supposed to—show your kids that they can learn anything and that every boring topic is fascinating when taught well. The Synthesis Tutor is not ChatGPT, but something more effective and personalized. It is custom-built with decades of DARPA research and has been proven across thousands of children. The last two years at Synthesis showed that we are better than anyone at how to teach advanced concepts to children.”

  Another company is Inspira Education Group. Inspira Education Group supports college and graduate school applicants with mentorship, counseling, and coaching. Wow, I wish I had this when I was going to my graduate school—that’s awesome.

  Using AI-enabled tools, the company pairs students with consultants from the world’s largest network of admissions counselors based on individual personality and learning needs. It actively incorporates AI into its core services. The company uses AI-enabled matching algorithms to pair students with consultants, and it uses AI tools to support counseling and coaching processes for college and graduate school applicants.

  We also have Panorama Education. Panorama provides a SaaS platform that gives educators tools and data they need to act on and improve student and school outcomes. It has AI-powered student support, utilizing leading multi-tiered system-of-support software to improve literacy, math, attendance, behavior, and social-emotional learning. It helps build students’ life skills with assessments, intervention tracking, and character surveys. Its platform measures factors related to social-emotional learning and student success in order to improve equity in schools and help educators improve core student outcomes. That’s really cool.

  And then of course, we have Rally Reader. Rally Reader was another episode on this podcast—you can go back and listen to that. Rally Reader offers an AI-enhanced e-reading platform for elementary and middle schoolers to improve their reading and literacy skills while engaging educators and parents with detailed progress dashboards.

  So it’s an AI-enhanced e-reading platform that acts as a personal reading coach. The AI provides real-time error detection and feedback to improve reading and literacy skills, and offers personalized learning experiences through word-by-word accuracy tracking and progress dashboards. Thanks, Mike. Back to you.

  Michael Peri:
  I mean, we could shout out one for our team here in Chicago—Consensus.

  Mike Collins:
  Yeah, great example.

  Michael Peri:
  Great example, really reimagining the learning and research side of education. That’s just one of plenty. So that list we’re going to share is just representative—I think the tip of the iceberg—and really will start to allow our audience to think about just how broad this can be. It could go from research, to creativity and book writing, to full-on curated ways to engage in learning that just resonates with the student at the end of the day.

  Mike Collins:
  And again, just to bring it home personally—I’m sure my son will love this—but I have a son enrolling at Columbia MFA program, and we talked for a half hour yesterday about how one can use this, even when you’re in school today, as a huge resource to make your work better. It is a tool. When I went to college, I had a typewriter, for God’s sake. Word processing was sent from heaven.

  And so it’s like the same thing: it is a tool. It is you and it—and you back and forth with it—making the work better. So again, I am short 529s and I am long AI and education and lifelong learning.

  Michael Peri:
  Yeah. And a call to action to anyone out there who’s building in this space—we often joke, like, “Oh man, the only thing you don’t want to invest in besides healthcare is education because of the sales cycle.” I think AI is going to be that breakthrough now. We’re going to see this be able to penetrate and distribute in ways that we didn’t think about—not even 12 months ago, let alone years ago.

  Mike Collins:
  I think it’s disruption theory, which is a lot of what people have been trying over the last 20, 30 years has been really trying to work with existing frameworks, and that just absorbs and treats it like an antibody and kills it. And I think AI is going to kind of circumvent that and it’s going to go right to the user and gradually become more independent as the child matures.

  But I think this one is— the time is right, and I think we’re going to see real big differences over the next 5 and 10 years in education that are going to be very transformative.

  Michael Peri:
  Thanks for bringing this one, Mike. This is one I think we could talk about all day every day, especially for folks with young kids thinking about this. So yeah, definitely a great topic.

  Mike Collins:
  I mean kids hopefully in your future, kids on the plate or grandkids potentially in the future, so it impacts all of us.

  Michael Peri:
  Exactly.

  Mike Collins:
  And if none of the above, it’s still a huge part of our society that we all live, work, and play in. What do you got, Mike?

  Michael Peri:
  All right, I’ve got an exciting one for us today—one that could really change the future of energy storage. On a personal note, my wife and I went down the rabbit hole on this topic because we finally upgraded to a plug-in hybrid car. So this breakthrough really caught my attention.

  Mike Collins:
  Welcome to the club.

  Michael Peri:
  We’re here and we’re loving it so far. It’s been a very nice welcome-to upgrade. So this is really about just reimagining a world in which batteries are not only more affordable and better for the environment, but also way more efficient. And I think the future is closer than we think.

  There’s some groundbreaking collaboration between researchers at the University of Chicago—shout out to my alma mater there—and UC San Diego. They’ve been focused on researching and developing a new sodium-based battery that could be a total game changer for how we store and use energy.

  For those out there that this might be new, why is this a big deal? Unlike lithium-ion batteries, which we currently rely on, they use materials that are super expensive and scarce.

  Mike Collins:
  Rare earth metal.

  Michael Peri:
  Yeah. With a sodium-based battery, it uses sodium. It’s super abundant, much cheaper, and it’s just a much more sustainable option, especially for large-scale energy storage. You have renewable energy grids—we invest a lot in that—and obviously electric vehicles.

  So they really leaned into this and said, okay, we have to think about how to tackle challenges that have held back sodium batteries in the past, because it’s not a novel idea for them to innovate here. What they did is they started to design a new anode material that’s optimized for this electrolyte.

  And early indications are it’s resulting in a battery that performs on par with, or even better than, traditional lithium-ion batteries. Plus, it works really efficiently at room temperature, which is a huge step forward. I think that was part of the big issue. So what does this mean?

  Mike Collins:
  How does it look at manufacturability? Because obviously, you’ve got to make this stuff at a huge scale, right?

  Michael Peri:
  Yep. So that’s it. I mean, this is where it’s like—so what does this mean for the future? It’s got to get off the bench of research. If sodium-based batteries can truly be scaled up and brought to market, it’s going to dramatically reimagine how we operate with this stuff today.

  It’s going to definitely reduce dependence on lithium, significantly lower costs, and finally make renewable energy more practical and widespread. And I think about that just in my one use case—electric vehicles. Way more affordable, much more accessible, and it eliminates the, “Okay, how do I have to think about this on a daily basis?”

  So it’s a cool one, super timely for us, and it’s always great to see some innovation out of USC.

  Samantha Herrick:
  This innovation that Mike has brought to the table today also has been published in what seems to be this podcast’s favorite scientific paper publisher, which is Nature Energy—and just Nature in general. They published their paper in Nature, and I’m going to read the abstract because I think that sort of helps round out exactly what we’re talking about here:

  “Anode-free batteries possess the optimal cell architecture due to their reduced weight, volume, and cost. However, the implementation has been limited by unstable anode morphological changes and anode-liquid electrolyte interface reactions. Here we show that an electrochemically stable solid electrolyte and the application of stack pressure can solve these issues by enabling the deposition of dense sodium metal.

  Furthermore, an aluminum current collector is found to achieve intimate solid-solid contact with the solid electrolyte, which allows highly reversible sodium plating and stripping at both high areal capacities and current densities, previously unobtainable with conventional aluminum foil. A sodium anode-free all-solid-state battery full cell is demonstrated with stable cycling for several hundred cycles. This cell architecture serves as a future direction for other battery chemistries to enable low-cost, high-energy density, and fast-charging batteries.”

  Mike Collins:
  And again, if you were to pick up—Kurzweil has really good graphs in his new book—basically showing all of these technologies and how they’re geometric real-time. Renewable energy just lasts 10, 20 years—90% improvement. Battery technology, 90% improvement over the last 25 years. These curves are all going in the right direction.

  Sometimes, they stall out and then there’s a step function as you have to change technology platforms. You’ve squeezed out the inner, but they just keep going.

  Michael Peri:
  And Mike, I feel like you hit it on the head. I think there is definitely consumer demand or pent-up demand, certainly enterprise demand. It’s just can you do all of this at scale and not disrupt current operations for an enterprise or the day-to-day for an individual? I think once we get through that, it’s going to be mass adoption almost overnight.

  Mike Collins:
  Yeah. And usually behind it, there’s some obsessive, crazy entrepreneur who’s going to build a factory to make the thing—like Elon did, right? So it’s a combination of the science, the lab, getting it off the bench, and then finding an entrepreneur who’s just going to drive it into the world.

  The last one today is the coming robots. And listen, let’s define robots broadly. We had a portfolio company, Ghost Robotics, just had a really nice exit doing stuff in clearing buildings instead of sending in the Marines. And then you’ve got obviously the huge factory robot—where I think Amazon now employs more robots than humans—and those two curves are on pretty fundamentally different scales right now.

  So, a billion robots on the planet by 2030—I’m using a broad definition of robots. But the one that just came out pretty recently that caught my attention and I think people need to be aware of is Figure’s 02 version. This is a company that is more in the humanoid-ish space.

  I know that Elon’s got his Optimus robot that keeps getting better with manual dexterity. His vision is one in every home. And everybody scoffs at that. But I’m old enough to remember when they scoffed at a desktop computer in every home. We scoffed at that. So again, we just keep seeing these improvements happening before our eyes—every six months, every 12 months—we’re seeing remarkable capability improvement.

  I mean, my car basically drives me to work now in full self-driving mode. I’m paying attention, I’m not reading the newspaper, and it’s at ungodly early hours, so it doesn’t deal with a lot of traffic. But basically, you push a button and it’s driving while I listen to a podcast.

  That didn’t exist two years ago, even though it was promised maybe five years ago. These things tend to be introduced, hyped, followed by disappointment, and then they become part of life. So I think everybody’s talking about AI, but the robots are coming too.

  And we know—we have portfolio companies doing this in food service. People can already see changes like, “Oh, now I check in my bags. Now I order my salad at a kiosk.” The next logical step is the robot will make your salad. I think by 2025, we’ll see high-volume retail shops with kind of a skeleton crew of humans.

  Samantha Herrick:
  Now, I don’t mean to freak anyone out, but this is already happening—not so much with humanoid robots, but retail stores are shifting to the skeleton crew that Mike’s talking about. One example is Zara.

  Everyone knows Zara, right? It’s a really popular shopping experience. They’ve implemented self-checkout in their boutique-style stores. They’ve revamped spaces to present a completely fresh image with new product display areas and efficient technology to provide better customer service. This includes fitting room screen-outs, dedicated spaces for online order collection, and advanced self-checkout areas.

  Here’s an example: a Zara store in the Philippines has really embraced this. Customers can browse a wide selection of collections, reserve fitting rooms through the Zara mobile app, and experience shopping in a completely new way. It flips what shopping is on its head.

  Speaker 7:
  Do you have a venture capital portfolio of cutting-edge startups? Without one, you could be missing out on enormous value creation and a more diversified personal portfolio. Alumni Ventures, ranked a top 20 VC firm by CB Insights, is the leading VC firm for individual investors. Believe in investing in innovation? Visit av.vc/foundation to get started.

  Mike Collins:
  And I think we’re within a couple of years of a consumer version of a family helper in the house doing stuff. So I want to call out Figure—check it out. This isn’t a portfolio company. Not everything we talk about here is shilling for our companies. But we do have others—Sam can put up a list—but robots are coming.

  And my final point: AI and robotics are going to increasingly be the story. We’re talking about both in one breath—it’s software and hardware. Combining the two is really hard, but when you do it well, you get incredible results—think Apple.

  One of the things we’re doing at AV is putting together 2025 funds, bringing robotics and AI funds into one because there are so many great opportunities that are one or the other—but even more exciting, both.

  Michael Peri:
  It’s super hand-in-glove. Early innovation tends to be laughed at or dismissed. Remember early robots? We had a vacuum cleaner going around our house and everyone said, “Oh, okay, it got wrapped around a cord, this thing sucks.”

  But as AI innovation skyrockets, it only makes sense that its deployment in robotics will be the ultimate showcase of what it can actually do beyond the commercial context. It’s easy to think about Amazon or government defense contracts—

  Mike Collins:
  Drones.

  Michael Peri:
  Certainly drones. Even in biotech or surgery, we’re seeing mass adoption in these large, regulated industries. If innovation is happening there, you know it’s just a matter of time before it shows up in everyday consumer life.

  Mike Collins:
  Yeah. One of my top 10 investments of all time was Intuitive Surgical. Early on, it just kept creeping into more and more applications. I think we’ll see a geometric progression—not linear—over the next 10 years for that kind of technology. There are so many benefits.

  Michael Peri:
  Yeah, no doubt.

  Mike Collins:
  It was a really exciting time. Another good episode, another good week. Thank you, Mike. We’ll do it again next week.

  Michael Peri:
  All right. See you, sir. Have a good one. Bye.

  Samantha Herrick:
  Thanks again for tuning into The Tech Optimist. If you enjoyed this episode, we’d really appreciate it if you’d give us a rating on whichever podcast app you’re using. And remember to subscribe to keep up with each episode. The Tech Optimist welcomes any questions, comments, or segment suggestions. Please email us at [email protected] with any of those, and visit our website at av.vc. As always, keep building.