Minter Dialogue with Jeremiah Owyang
Jeremiah Owyang is a renowned industry analyst based in Silicon Valley, who has a beat on emerging tech. He’s a General Partner at the Venture Capital outfit, Blitzscaling Ventures and a sought-after speaker. In this conversation, we discuss the explosion of AI, how he views and evaluates AI startups, to what extent ethics and transparency are used or necessary in these AI-infused initiatives, the humanisation of AI, the rise of therapeutic AI and much more. A fascinating update on where AI is going, including some hot new apps you need to check out.
Please send me your questions — as an audio file if you’d like — to email@example.com. Otherwise, below, you’ll find the show notes and, of course, you are invited to comment. If you liked the podcast, please take a moment to rate it here.
Further resources for the Minter Dialogue podcast:
Meanwhile, you can find my other interviews on the Minter Dialogue Show in this podcast tab, on Megaphone or via Apple Podcasts. If you like the show, please go over to rate this podcast via RateThisPodcast! And for the francophones reading this, if you want to get more podcasts, you can also find my radio show en français over at: MinterDial.fr, on Megaphone or in iTunes.
Music credit: The jingle at the beginning of the show is courtesy of my friend, Pierre Journel, author of the Guitar Channel. And, the new sign-off music is “A Convinced Man,” a song I co-wrote and recorded with Stephanie Singer back in the late 1980s (please excuse the quality of the sound!).
Full transcript via Otter.ai
SUMMARY KEYWORDS: ai, jeremiah, companies, people, minter, pi, silicon valley, years, reid hoffman, part, hear, startups, ethical framework, tech, empathy, happening, google, crossfit, san francisco, convinced
SPEAKERS: Jeremiah Owyang, Pi, Minter Dial
Minter Dial 00:05
Hello and a very welcome back to the Minter dialogue and a happy new year for 2024. This is episode 550. My name is Minter Dial and I’m your host for this podcast, a most proud member of the Evergreen Podcast Network. For more information or to check out other shows on this network, go visit evergreenpodcasts.com. So, this week’s interview is with Jeremiah Owyang. Jeremiah is a renowned industry analyst based in Silicon Valley, who has a beat on emerging tech. He’s a general partner at the venture capital outfit, Blitzscaling ventures, and a sought-after speaker. In this conversation with Jeremiah, we discussed the explosion of AI, how he views and evaluates AI startups, especially from a VC standpoint, to what extent ethics and transparency are used, or necessary. In these AI infused initiatives, the humanization of AI, the rise of therapeutic AI, and a whole lot more. It’s a fascinating update on where AI is going, including some hot new apps you need to check out you’ll find all the show notes on Minter dial.com And if you have a moment, I urge you to go over and drop in a rating and review. And don’t forget to subscribe in any event to catch all the future episodes. Now for the show. Jeremiah Owyang, so great to have you on my show. Frickin’ forever have I wanted this, you and I have paths that have crossed many times over the years I’ve lost touch except I see you in the press all the time. You are an icon a beacon of what’s going on in AI right to have you on my show, in your own words, Jeremiah, who are you?
Jeremiah Owyang 01:55
Hey Minter! I loved digital technology trends, my role changes from time to time. But I have a career mission, I help companies connect to their customers using new technologies. And that’s really been my theme. Now I’m a venture capitalist, I run an AI fund investing in startups.
Minter Dial 02:16
You must have a lot of options to be investing in.
Jeremiah Owyang 02:22
This market has just exploded in this 2023 period. When I started tracking the market, there’s a website called theresanAIforthat.com. Of course, there was 3000 projects. And today, there are 10,000 projects at the end of 2023.
Minter Dial 02:42
How to sort through them and figure out which ones which ones which? I mean, I suppose you you’re looking mostly for gains in finance. But do you have other things you’re looking for in terms of moving the dial.
Jeremiah Owyang 02:55
So, most of these companies are pre-revenue. So, finance is not always the way you measure but you look at growth rates, adoption rates, what types of partnerships they have, mentions in the press, traction; but we also look at their business models to see if they have a business model that will grow very fast.
Minter Dial 03:18
I want to get into that a little bit later. But Jeremiah, just first talk a little bit more personal. I know, one of the things that marked me in my knowledge of you and the journey that you’ve been on, is you’re embarking on CrossFit and being a total freak of fitness. Uh, tell us how that’s marked you and why it’s so important in your life.
Jeremiah Owyang 03:42
Okay, thank you for asking. You know, that’s one of the things that is joked about never to talk to anybody about CrossFit because they won’t shut up about it. So, just want to warn you, you just pulled the plug. Okay. So, I when I turned middle age, which is 40 for most people, which was several years ago, I was out of shape. Because I’d been starting companies. I couldn’t run a mile. I had mild health problems. I was overweight, and I couldn’t even do a pull up or run. And so, I decided to fix all that. And several years later, I’m hitting all the marks fixed all those mental and health issues that might have been plaguing me, that I could, and now I’m quite fit. And go to fitness CrossFit at 5am, three to four times a week. I went this morning at 5am. We did deadlifts and I hit a personal record which means, you know, achieving something. So, I’m getting stronger. Even though I’m approaching half a century I’m getting stronger. And it helps me with having mental fortitude. Because you do something every day that’s challenging and you fail every almost every day that you try this. And you’re with people who are very positive. If you do this thing, this very positive thing. Now all of this I trained for obstacle course racing. So, there’s a there’s an organization called Spartan Races, where you run in the mountains in the hills. You And you climb over walls and you carry heavy things and you throw spears and you climb up ropes, and you swim like water, mud. Oh, yeah, that’s the, that’s the benefit. That’s the fun part because it’s nature. Listen, we are so detached, unattached from nature. We’re about to talk about AI. So, what’s the opposite of Silicon Valley where I live in tech and AI and social media. It’s rolling in the mud. Yeah, mud mud. But that’s Earth. That’s the mother. So, there’s something purifying and rejuvenating about pushing yourself when you’re close to Earth.
Minter Dial 05:37
One of the books that really made an impression on me, and I think like you, I’ve also been looking at AI in a therapeutic sense, but is a book by Johann Hari, and it’s called “Lost Connections.” And, you know, we’re all so connected, but yet so detached, as you said, and Johann talks about the seven ways to reconnect. And of course, one of them is putting your hands in the dirt, and getting reconnected with nature, something we should all do a lot more of. So, thank you for sharing that. I think it is something that we all ought to be thinking about, as much as we’re about to, you know, promote and talk about the greatness and most amazing things and also dangers of AI. But we should also remain humanly connected and character the back into Earth. Whilst that so one of the things that people are most I mean, there’s basically, unless you’re living under some kind of ridiculous Stone, no one’s now knock out the plot that AI is a big thing. It’s it become thanks to open AI and on the lambs, a topic amongst 70 to 80-year-olds. I mean, I’ve been to dinner parties with octogenarians who are all excited to talking about it. So, everybody knows about it. I would say, I don’t know if they know it correctly. But for you who are in it, I get the feeling that it’s become so big, it’s hard to really actually stay abreast of it.
Jeremiah Owyang 07:11
I wouldn’t say it’s becoming so big, it’s just moving so fast. I think that is the thing. I we haven’t seen a market move this quickly with that amount of startups growth and growing birth so quickly. So, I’ve been in Silicon Valley for 27 years. And so, I was here as an intern. In the.com. Era, I was in.com, web two sharing economy, web three. And now AI, so five trends. And I’ve never seen anything move this quick. You know, for example, we already mentioned the 10,000 startups, which has tripled since March. I mean, that’s incredible. Every single large tech company, all of them are now AI companies and positioning as such, from Salesforce to SAP to, to all of them to Google. Up and down 101, the main thoroughfare in Silicon Valley, the freeway highway 1/3 of the billboards per my count are about AI. That’s quite unusual. And in San Francisco, which has its fair share of problems of crime, and then there’s vacancies in commercial real estate. There’s 100 ai events per month, mentor, that’s three AI events per day. And on Saturday, that’s two in Sunday, that’s two in person 100 ai events per month, that’s never happened at that level, that rate. So, something is happening. And it’s very different, let alone the adoption rates. Open AI, which the version that we know GBT birth on November 30 2022. And a year later, 100 million users adopted it. That’s perhaps the fastest gross growth of any app that we’ve ever seen. Let’s put Threads aside because it was just an outshoot of Instagram. It’s not quite a fair comparison. So, this is something big.
Minter Dial 09:09
My point was actually how AI is now being used everywhere. Yeah, that’s, that’s why it’s sort of wide. It’s, you know, if you want to stay up to date with AI in let’s say, marketing, that’s one thing. But as AI in medicine, there’s AI in robotics or factory management. There’s so the level of deployment is hard to keep up with. There’s the speed on one end. But there’s the breadth of utilization now that C makes it so hard and it’s almost everywhere.
Jeremiah Owyang 09:44
And I don’t think has been fully integrated everywhere yet, but it’s starting to integrate. Yeah, I mean, it’s just where we’re at the first year of generative AI at from mainstream adoption. I think there’s a lot more to come. We’ll know it when it’s really here. When we are all at inbox zero with your emails, and I don’t think that’s far off Minter, I think that’s like a very common use case, we’ll see that Google has to lead. And Yahoo has to lead. And outlook from Microsoft has already said they’re going to do that. Because the training data is already there. It’s already known problem, people are willing to pay for these things. So, inbox zero because AI, reading, sorting, summarizing emails, and then starting to reply on our behalf. I think that’s just 24 months away, or less. So
Minter Dial 10:33
I want to push into that one second, because I remember a former colleague of mine, the number two in my division, and he would say, Well, I have Inbox Zero every Friday night. And I was like, really, he says, Yeah, I just do a select all and delete. And if they’re important, they’ll come back and do it again, right to begin next week. But until then, it’s zero. I don’t need them. And I was like, well, that is one kooky way of doing things. And then when I listened to you, in this manner, where the AI is doing it for you reminds me of people who outsource their personal branding. You don’t know what they’re saying about you anymore. How do you stay on top of the content? You know, if you if an email comes in saying decision make an easy decision? Should I buy 10 pencils? Okay, well, you can let the AI do “I should buy 10 pencils,” but then you can’t be cognitively aware of everything that the AI is doing for you. So, how does that get worked out?
Jeremiah Owyang 11:36
Well, we’re not able to be cognitively aware of all of those emails that are already in our inbox, hence your friend deleting everything. I mean, Minter. How many unread emails do you have across all your inboxes, right now?
Minter Dial 11:48
Yeah, plenty. With Gmail, doing that great filtering and moving into social promotion and so on, and the spam, I’m probably down to 30. On average, in my personal inbox.
Jeremiah Owyang 12:09
Yeah. But across all inboxes. Right, it’s just not even possible, we sign up for all this stuff. So, how do you know we’re not missing something important? For example, you and I were in an email conversation last week, and I didn’t see your email because you send it to an inbox, which I don’t use normally. So, that’s already a disconnect. It should already have flagged that you are high, high profile and high importance, which you are. So, okay, I think the problem set is very clear. So, here’s the interesting thing about email. There’s already a tremendous amount of data, 20 years of data of a historical data and behavior. I mean, it knows how we respond and who do we respond to which email? Do we open quickly? Which ones do we ignore? What shouldn’t we did we open four times? And then finally respond? How did we respond? It’s possible, though, that if there’s a decision to be made, that the AI agent, the email agent, should flag a summary and says, there’s three decisions to be made today. Here they are. They’re prioritized for you to make your decision, so you can get back to your CrossFit workout.
Minter Dial 13:11
And presumably, like much like Pega, for example, the AI would further say, Listen, these are three options you can use for the decision making with benefits attached to it, and such and such. So, you have that final agency, oh, I’m still doing something before my CrossFit. Jeremiah, when you are evaluating the VC, just to go back to that. you’ve got a huge number of choices. And obviously, many of them aren’t doing money. But what are the criteria by which you really jump in on them? For example, we used to say in the old days, while I don’t care about the idea, I invest in people, or is it going to be, you know, mostly about the business model you say? How about their ethics? What are the things that you flag as as no goes? And what are the things that say, oh, gosh, I’m all in.
Jeremiah Owyang 14:05
Gotcha. So, I have been an angel investor for six years, but I have not been a formal general partner at a VC firm until recently. So, that’s a new appointment that I’ve that I have stepped into. So, I’ve I have joined Blitzscaling ventures. Now Blitzscaling is based upon a book that perhaps you’ve heard of by Reid Hoffman and Chris Yeh. And this is a well-known firm in book. So, Chris is the founding partner and my friend for 20 years and I served on his board of a company he sold. And coincidentally I’m an LP, which means investor in the fund, in the first fund. So, I have a long relationship with Blitzscaling ventures and Chris and Reid Hoffman. Yes, that Reid Hoffman, Pay Pal, LinkedIn board of directors, Microsoft Open AI, and now Inflection AI. He is one of the advisors and LPs. And so, our investment thesis to answer your question is based upon the book Look, which is based off of Stanford class, which they interviewed the leaders from web one (dotcom), web 2.0 and the sharing economy to find out what works. So, we’re looking for those types of criteria, for example, we’re looking for, are they in a market where there could be a winner that takes most where it’s a rapid growth market? Two, we’re looking for companies that have network effects, whereas every user that participates adds value for other users. And number three, does it have viral effects. In Silicon Valley, that parlance is product-led growth. And then the fourth one is a distribution strategy that, for example, Open AI has Microsoft to help them spread their product throughout their 365 Suite. Now, to the excellent point that you brought up, evaluating the founder is really important. And so, we have things we look for. No formal scoring for that one, we do have quantitative scoring for the other ones I just mentioned. But for the founder, we’re observing their behaviors to see. Can they learn, are they an infinite learner? And can they pivot? Do they change? What’s their pedigree? What have they done in the past things like that?
Minter Dial 16:08
What about the ethical framework? With AI, there’s a lot of talk around ethics, it strikes me as maybe something that’s more likely to be imposed on a large company than a startup. I mean, most startups are still sort of behind the eight ball when it comes to their brand and all the other things, they are just trying to get the bloody product out. But the ethical framework? And a lot of people talk about it, but I feel like ethical frameworks are, at best something we talk about. And worst, probably we have no idea about.
Jeremiah Owyang 16:47
This is a sticky topic. So, we are Silicon Valley’s certainly paying attention to the news from the EU on the regulations put forth and we’ve looked at it. Recently, last month, the White House put forth an executive order with some parameters around what AI regulations could be. In both cases, these are all proposed and with EU would win take place for a number of quarters. Now, to your point, many of the startups who I’m mostly focused on they don’t know if they’ll survive till in two quarters. They literally only have that much runway. So, to task them, like are you aligning to future regulations that are not yet and legal requirement? It? No, it’s not. However, the larger AI companies such as Open AI, Anthropic, whatever Microsoft, Google and Amazon are doing, yeah, they have to pay close attention. In fact, they’re helping to shape those things. So, we obviously are looking for, we’re not going to be investing in a founder that has a criminal background, or has been accused of misconduct in in a way that has been proven to be true. So, but there’s only so much we can evaluate on their ability to be ethical. By the way, there’s one way we can find that out. Are they fudging the numbers when they’re trying to pitch VCs? And are they promising things to VCs? And when we check with our VC friends, or their customers, or their partners, and we find out it’s not true, it’s we can quickly find that out whether or not they’re following AI ethics would be a secondary.
Minter Dial 18:23
Yeah, I mean, of course, probably the fudging goes on in some sort of formula in your ethic in your Excel spreadsheet. But going back, for me, the ethical framework isn’t necessarily around regulation or legislation. But it’s more about doing good or not doing evil per se.
Jeremiah Owyang 18:42
Oh, purpose. Okay. So, that’s so I, when I hear the term ethics, and I think many people in Silicon Valley would agree, it could mean a lot of different things. So, in the way corporates sometimes use that in US side, it’s around. Are you following the laws and generally agreed upon social behaviors? That doesn’t necessarily mean are you trying to do good, which I think is a different level up. And that doesn’t come up as often. But often, we see many of the tech companies and AI foundational models, talk about AI for humanity.
Minter Dial 19:24
AI for good AI for humanity, big topic for me, AI for humanity. I just wrote a 3000-word article on the topic of the humanization of AI. And I certainly want to ask you about where we are on that. Do you do you think we can do you think we should be trying to humanize AI?
Jeremiah Owyang 19:48
Yes, so AI is trained already on human information and data. Even our tweet exchange this today As I put out a tweet asking people would you want a digital avatar for you to last everlasting, post death? And you had a fun retort that said, Only if I can be involved in the pre training, which was very clever. Because it means yes, you’re teaching it yourself. It’s based on you, of course, I think that’s a requirement. So, right now open AI and inflection anthropic, they’re using data that was built by humans. But that won’t always be the case, right? Because soon there’ll be synthetic data that is trading off, and that’s happening already. So, yes, it needs to be more human. And I’m sure it’s the case in your region, but loneliness, anxiety and depression are at an all-time high, that’s ever been measured in the United States, the highest levels, it’s never been higher, and loneliness is at a very high rate. Now listen, I can hear what people are saying, in groaning and rolling their eyes. The best thing is to be with people, the best thing is to be with loved ones. But for many of them, they can’t, they don’t have the people around them, or they can’t afford mental health professionals, there’s just not enough or they’re not available at 3am. So, AI is going to be a surrogate in that way. And it’s already starting to happen. I mean, it happened with Woebot 10 years ago. And then we saw Character AI, then we saw Replika which had its issues. And now we see other versions, which we can talk about, from Inflection to Open AI and to MIRI and others.
Minter Dial 21:31
And to Pi.
Jeremiah Owyang 21:33
Yes, which is, which is part of Inflection.
Minter Dial 21:35
Oh, I see. Well. So, let us dig into that a little bit more, because it is an area of deep interest for me and I certainly with the eye-rolling comment, believe that we ought also to be thinking about why we have such high levels of anxiety and depression, and loneliness in society. I mean, to what extent tech is participating in that it does seem to correlate with the more there’s tech, the less there is connection, the less we have friend, deep friends, I mean, I’m not saying it’s tax fault, but I do see a correlation. And, and therefore I feel like we need to have a little bit more intention, about going back into the mud, going back into our friends and hugging. And knowing how to fight without running away like astronauts in a capsule, where you actually have to figure out how to get through hardship where you actually have to learn to to know how to disagree with one another. What do you think?
Jeremiah Owyang 22:39
I think you’re absolutely right, that technology has created a divide. It’s helped us to connect loosely. It’s created to divide. But it’s helped people to hear things. And it has distortions, too. So, Minter, with every technology and tool from fire to blades to obsidian to splitting the atom, it has and will be used for good and bad. And it’s up to the humans who wield it. Of course, there’s responsibility in those who create those technologies to help them guide them towards humanity. But we know every technology and tool in the past has been used for both. Now, interestingly enough, so you’re in London, I’m in San Francisco. Prior I mentioned there’s 100 in-person AI events. Some of them are at houses, mansions, event centers, restaurants on the beach. And the interesting thing about San Francisco is it’s a fist bump and hug culture. So, you first bump or you shake hands with people you don’t really know you don’t see that often. But once you see them a few times, it’s a full-on hug, embrace. And that’s part of that, that that hippie culture, the Burning Man culture, I mean, the hippie movement, 69 in San Francisco, it’s like, yeah,
Minter Dial 23:50
You see my iPhone cover?
Jeremiah Owyang 23:52
Yeah, it’s the Grateful Dead. Like, the amphitheater where they play is just, you know, a few minutes from me, you know? Yes. So, you know the culture. That openness and that connectedness. So, yeah, that’s part of the culture. So, why would 1000s of AI builders, why would they want to be in person? Why would they want to be at a beach together in a bonfire? Where there’s no computers? Yeah, we on our phones. Why? Because the culture and the ethos is something unique in in San Francisco, we have the culture called AI first. Now, you probably are saying, Oh, well, what about humanity? First, well just hear this out. So, the AI first culture in Silicon Valley is if you have a problem, then you try to get AI to do it first. And if AI can’t fix that problem, then you build it. Then if you can’t do that, you hire somebody to build it and then you got a company. And the end goal, of course is so we have more time to do the things that we love as humans. So, that’s kind Have a subset of the culture that’s not well said I am quoted in the New Yorker on that exact quote that I just gave you. You know New York likes to shit on San Francisco so I hope they heard that.
Minter Dial 25:14
Poo on you too! Obviously when you talked about the number of meetings you’re having in San Francisco, you and I, we speak for a living as well, and the nature of conventions and conferences — in London anyway — it’s very difficult to get bums in seats. People would just assume you know, listen to it without having to one-and-a-half-hour commute with smelly, stressed people in the in the train, and then rushing through the rain to get to the place. They’re like, well, I can just do it from my country home. And that seems like a lot better than having to deal with all that. And yet, you were talking about, you do a lot of events yourself, your Llama, your Llama event and I don’t remember the exact word you use. But it’s I mean, it was in the lounge, crazy. The Llama Lounge. Bananas, Yes!. There you go. Great. But yeah, and so these, these are geeks of the highest order who are curled coagulating around. And I wanted to get back to this idea when he said something started you started your career in at Hitachi in community.
Jeremiah Owyang 26:26
Wow, good memory.
Minter Dial 26:27
And I want to link that into this idea of loneliness, and detach this. And I wonder in Jeremiah’s mind, to what extent community is less, equal or more relevant than it was back in that day.
Jeremiah Owyang 26:45
More important, especially coming out of COVID, where people were completely severed, which helped to trigger a lot of depression, anxiety and sadness and loneliness. Some people are still struggling to come back to physical events per year, just comment on those who left the city to probably buy their country home. And that happened here as well. So, yes, people still want to be connected to each other. Some people still are unsure whether they’re extroverts or introverts there, there’s a reevaluation happening now. And these physical events that the tech communities doing as part of that, AI and social media are part of that, both positive and negative.
Minter Dial 27:33
So, I want to lean back into this idea of community and the zero inbox, maybe link it into dating. Not that I am dating, I’m happily married, my daughter has spent some time on her thesis about dating apps. And there are certain apps that going back to your idea of AI first, that will help find a date, engage with a date, have the first three or four filters and then you can say, Alright, now we’re ready to meet. I feel like that is a disassociation, because at the end of the day, even small talk is part of how we connect. And so, when you are evaluating AI, to what extent do you have a personal thinking about these things? I mean, in other words, if you say that is not something that I would do? How can do you detach yourself from that? Or do you bring that element of you into the decision making?
Jeremiah Owyang 28:34
It’s you have to be careful about bringing your view only because as an investor or prior, my role was an industry analyst, you need to encompass all the views. That’s really important. That’s very important. Otherwise, you’re going to end up making very poor choices and analysis and decisions and investments.
Minter Dial 28:54
But we’ll see if you go back to this idea, I get that, of course, that this might not be good for humanity. Okay, that’s different. Well, that’s what I’m getting with that is that if you have an app, which really is scalable really is obviously needed because people are lonely. People have struggled struggling today. Well, let’s just allow all that to be easy. And make that the way we sold and for that on the side make a lot of money.
Jeremiah Owyang 29:23
Yes, so the one that comes to mind right now and I’m getting bombarded with ads on this and Instagram is AI Girlfriends. And that’s troubling to me. But I can see why so many disenfranchised, dislocated mentally, emotionally sexually immature men are purchasing those apps. I didn’t say boys I send them in. It is an outcome of society. But as what they will do is they’re starting to form emotional relationships with AI girlfriends, and you can choose your girlfriend the way they look. And they can be caring to sexual and, like, I see that as it’s all over Instagram and Tiktok. It’s hard to tell, by the way now, what’s human and what’s not? It’s really hard to tell. It’s really challenging to tell now. So, there’s definitely downsides to that. But at the same time, is it scratching the itch that these or gentlemen are, and women to I’m sorry, let’s, let’s go on both sides, that they are truly lonely, and there’s a void that needs to be filled. This is something that’s going to happen.
Minter Dial 30:42
I agree. I mean, replica was an exact another example of that, where you can have, you know, a far more sexual relationship with if you choose to. And I mean, they tried to turn it off, but there was so much demand for it. I mean, again, it does point back to the issue of society itself. And on notwithstanding the option, or the leader see the opportunity that these provide much like in therapy, and I’ve been looking at Hume.ai, and Mpathic.ai. Got here. And there are numerous therapeutic AI ventures that aren’t there. Yeah. But going back to this humanity story, I do want to have a little bit more from you on to what extent do you think things like empathy, creativity, intuition, and I’m not going to say consciousness or emotion because they aren’t embodied. But for the others…?
Jeremiah Owyang 31:33
They emulate it. They simulate it. Okay, so let’s broaden beyond interpersonal relationships. So, let’s talk about friendship therapy and coaching, because that’s the use case that I see. And we’re going to be investing in that category for sure. So, we already see the LLMs, they can act in that way. You can ask it to act in the way of a therapist give an answer, but GPT doesn’t have a dialogue. Another one, which is Reid Hoffman’s company, and we’re tied to Reid Hoffman is called Inflection and they make Pi. And you had a chance to trial Pi? I believe. I did. Yeah. Be curious to hear your thoughts. So, pi is stands for Personal intelligence, and Pi has a dialogue, you can leave the microphone on. And it’ll have a full-on dialogue, like you’re on a phone call with another human being. There are six different voices to choose from. I found voice number four to be very realistic, or West Coast woman. Number five, and six have pseudo-British, Queen’s English tones to them. Be curious, on your take on that. And you could have a dialogue? And it will say, Oh, I’m sorry to hear that. Have you thought about this? And it’ll, it’ll provide a replica of what empathy is like and have this dialog and help you walk through and talk through things. Eventually, it’ll connect to your systems in your house, it can be your personal assistant help with your scheduling your shopping. So, it’ll be more than a coach or a therapist or friend, it can be many different things. But these types of use cases already here. So, I’d love to hear your take on how did it go trying Pi, which is less than a year old?
Minter Dial 33:10
Well indeed, So I haven’t explored it to its degree. But I did ask it to in it because I looked at the productivity side because it has different functionalities. Right? I getting it right was that Perplexity. I can’t remember.
Jeremiah Owyang 33:25
Perplexity is a search tool.
Minter Dial 33:27
Oh yeah. So, this was with Pi. That’s right. I like you, I try to try out stuff. And I try to figure out what, what I liked about it. I asked it to make for me engaging tweets around an article and I just gave it the link to the tweet. So, it clearly read the article, and then produce to me for me tweets. And so, I admonished myself in not making a better prompt, which is also such an important part of these things as you start learning the model. But I had an agreeable experience, and I was, a little were fine-tuning, as I think it’s always generally still better AI plus human in terms of effectiveness. So, I adapted a few of the tweets, and I buffered them all out. So, here’s to Pi.ai.
Jeremiah Owyang 34:16
Yes. Did you try the voice features and have a full-on dialogue with it?
Minter Dial 34:19
Not yet, but I shall now.
Jeremiah Owyang 34:23
So, I want to demo it, but I’m not able to because I’m having those audio issues. Oh, right. Yeah.
Minter Dial 34:28
Is there something you maybe you can send me a little something I can show a or put in the show notes?
Jeremiah Owyang 34:33
I think you should try it and record it yourself. So, yeah,
Minter Dial 34:37
I’ll do that. I will do that. You have you have now sent me a challenge, sir. Challenge accepted.
Jeremiah Owyang 34:44
It’s possible I could show an example. But I’d have to leave the first Zoom account. Is that possible?
Minter Dial 34:52
Yeah, sure. Try.
Jeremiah Owyang 34:53
Okay. So, I’ll stay on this second mic so you can still hear me right? Yep, absolutely.
All right, I can’t code in Python directly.
Jeremiah Owyang 35:02
Alright, here we go, Hey, Pi, my dog died. And I’m feeling sad. What should I do? So, I’m writing that in. I could of course, use voice if I wanted to. I could use…
I’m really sorry to hear about your dog’s passing, losing a beloved pet can be incredibly difficult and painful. It’s completely normal to feel sad. And there’s no one right way to cope with grief. Here are a few things that might help. Give yourself time and space to grieve. It’s okay to feel however you feel. And you don’t need to rush through the process.
Jeremiah Owyang 35:35
Ok, I going to put it on pause. But you were able to hear that right?
Minter Dial 35:38
And the sound of the quality of the voice. The intonations, the inflections were absolutely stunning.
Jeremiah Owyang 35:47
That’s right. So, if you haven’t tried Pi, and you’re focusing on AI and empathy, you’re missing it. Like this is the one that is going much deeper, you can ask it about death, divorce, Love marriage, getting along with friends, colleagues, like it is simulating empathy. No other AI is doing anything like this. This is something else.
Minter Dial 36:09
Well, I love that you’re telling me about that. As you know, this is a deep topic for me. Can you without spilling the beans? Let us know how it’s becoming more empathic. Is it really a question of the dataset or the tagging, processing power? What’s going on to make that empathy more powerful?
Jeremiah Owyang 36:29
Well, there’s some London ties here. So, there’s two co-founders, one of the co-founders, Mustafa Suleiman, who is from who spent with Google DeepMind, which is from London. And he left Google to start this and he had his co-founder Reid Hoffman, yes, that Reid, LinkedIn, Pay Pal, Microsoft board of directors of Open AI now who stepped back from Greylock and stepped back from Open AI. He’s the co-founder of Inflection. And Mustafa and Reid Hoffman is an incredibly powerful duo. And they are creating this amazing technology. It’s a foundational model and LLM. So, they’re training it obviously Mustafa knows how to train things they built deep mind, they built AlphaGo, right, that beat humans that go, which was unbeatable, right? So, this team can do anything. And they hired, I believe, 50 ethicists, to help train and to guide and advice from psychologists to therapists to athletes to people who focus on regulation. Reid is counseling, government leaders and religious leaders, he already has connections to those that are making the laws. So, they already understand these things. And if you read, read, reads book, which is called impromptu, he talks about how AI can help humanity and goes through multiple use cases. Also, Mustafa published a book two, I think it’s called the coming wave, which talks about how AI is going to change the world and also calls for regulation, requesting regulation in the book. This is, in my opinion, these guys are on the right side of history that are trying to tie this towards to be aligning to help humanity. So, I’m very excited, there will be an API. For those that are not technical, it means that you can connect other apps into pi in the future. So, if it’s your home automation, or your email or your podcasting tool, you need somebody to talk to you during the podcast, you have no guests. someday it will be it there will be Pi.
Minter Dial 38:35
Maybe I should just do that for the hell of it. So, Jeremiah, we’re talking a lot about tech per se, but a lot of companies aren’t in tech. Let’s say you’re not a manufacturer of face creams, a manufacturer of ladders, an oil and gas company. Is there any industry that shouldn’t be saying at this point? Oh, shoot, I need to be having a chief AI officer on my board? Or do you even recommend that as an approach for most companies?
Jeremiah Owyang 39:16
I can’t think of an industry where AI is not impacting it. I cannot. Well, I’ve already met startups that trained MLMs on the Bible and the Koran already. Just think about that. But by the way, one religious DD talking to another AI avatars like that’s starting to happen. That’s kind of a wild thing. I can’t think even like the forestry services, they need Internet of Things and sensors to track the health of forests and to prevent disease or wildfires or orange and when you know as climate becomes more extreme like AI can help them with those things. I can I cannot think of an industry that doesn’t have it. Spirituality already has AI projects. Even the Vatican leans into tech, usually about a year or two later, but it’s they do over time. They’re quite progressive. I can’t think of anything that is not using or being impacted by AI. Whether or not they have a chief AI officer, I think varies whether or not they’re going to lead it in as part of their product in their business. I don’t think it’s necessary for every company to have a chief AI officer. And I would have said the same thing about the chief Metaverse officer before that.
Minter Dial 40:33
Yeah, I have right beside me, church times from September 2023. And there was a webinar called AI and the church a church times webinar with Beth Singler and talks very much about this thing. And ultimately, what that brings to my mind is the very existential nature of AI. And, and how I’m guessing you also have the same thing still, where you talk to an audience and someone’s going to raise their hand says, gee, are you scared of the potential of AI? Or, you know, should we be all worried and, and be, you know, becoming AI coders? But there is something very existential about it. About consciousness or death? Your question about the avatar, suggesting it can live on after you die? Yes. The issue for me, Jeremiah, and something that I push into is I think that it’s very easy with tech to be dissociated from reality. For example, the fact that we die. And there’s a school of thought that says, Oh, well, aging is just a disease, let’s get rid of it. And let’s defy death. And I feel like that’s part of the problem we have today in society, which is where evacuating reality for some sort of artificial life that ultimately isn’t fulfilling because life is about dealing with reality, the shit, the slog through the mud in your obstacle course, and the loss of somebody, this is what life’s actually about. And there’s there shouldn’t be an effort to hide from it or to revel in the artificiality of it.
Jeremiah Owyang 42:20
I generally agree with you, I think it’s healthy to think about death frequently. I went through a lot of studying on the stoic philosophies when I was having my own mental trials, emotional trials. And that brought me to that point, and I’m okay with it. So, I think that’s, that’s part of that. And it reminds us to do our best in the very short period of time that we have. Now, going back to the profound thought that AI has and the existential opportunity, if not threat for humankind, I think we should be very, very clear that we are building a new species. That’s essentially what it is. And that has amazing upsides and potential risks. It can see with GPT, for vision, it can hear with whisper, which is an API plugin, it can think with an LLM API plugin, you can speak because of the LLM. It can simulate emotion as we just gave a demo, it can stimulate Marcus Aurelius if you prefer to do so. So, all of these things are now it’s like a subspecies of humankind. I think that’s really what’s happening.
Minter Dial 43:36
Yeah, there’s a podcast, right of have a little bit of a mind fart, of remembering who actually said it. But they talked about the synthetic relationships that we have, where we are already developing that with our phone, we’ve always been doing that with machines in the past, but we now have these opportunities for synthetic relationships with AI bots, and so on. I live five days Jeremiah with a, an empathic bot 24/7. And I can tell you, it was a profound experience. And I don’t think it was artificial. Let me in the last question I have for you, Jeremiah, I think I would love for you to riff on this notion of transparency. So, just the context of it is that if you have an AI, it does seem like a black box a lot of the time. And there’s regular requests for more transparency, more transparency of how a company is more transparency about the secret sauce in what they’re doing. And then transparency in AI. Oh, how do you approach transparency and to what extent do you promote it with the companies for example, you’re investing in.
Jeremiah Owyang 44:52
Okay, so there’s a there’s a conflict here. So, I read the EU regulations requesting model transparency. It’s a unclear if those, the actual requests will be in private with the IP, or will be a republic document. And unfortunately, many tech companies cannot comply with that because it is their secret sauce, it is their IP. And there are certain things that they will not be able to be transparent with how the algorithm works. And there’s actually very important reasons for society that you do not want that algorithm to be public. For example, if Facebook made it public on how their algorithm for their news feed, or Google news reader was public, or if Twitter was public about exactly how it works, and or AI, then it can be gamed by nefarious parties. So, and secondly, it’s going to be changing all the time, it should be adapting and learning and modeling. So, there it is the conflict to request transparency of the tech algorithms, whether it be AI from or to a news feed. And I think that’s a real issue that we’re all going to have to contend with. I don’t know what the right answer for that is. But I do believe that Silicon Valley will generally resist that.
Minter Dial 46:05
Yeah, I tend to agree with you. And I fundamentally believe, Jeremiah, that there’s always an agenda behind a request for transparency. And that agenda is not necessarily good for business. It may not even be necessarily good for society.
Jeremiah Owyang 46:22
That’s right. And there’s been numerous German, and now French LLMs that have emerged in the last few quarters, which could be competitive to the Silicon Valley LLMs. So, that there will be questions on where’s that information and secret sauce going to go?
Minter Dial 46:38
Indeed, and when you look at a company in this neat, neat notion of investing in your venture capitalism? Yes. What about the intention of the company? Is that something that’s relevant? I mean, I, we talked a little bit about purpose before but just the intention of the company, does that set? Is that a space that you investigate? What are you trying to do in a in a more in a grander scheme, or is that something that’s just it’s about making more money or making success?
Jeremiah Owyang 47:16
This is a longer conversation. So, the larger tech companies in particular, the foundational models are talking, and I mentioned this prior talking a lot about helping AI to improve humanity. And in the end, if you do not align your business to helping humanity, you will be in a big problem. For example, Facebook, lost their way. They were about serving the people who were generating ads more and same with Google Ads is over 50% of the revenues, and not necessarily serving the end users. And this is why people are moving away from both of these companies in some regards. And so, if you’re not aligned with humanity, then your business will not sustain and therefore you cannot keep going. So, that is a second or third or tertiary thought in the things but the primary objective for companies is to sustain through ongoing growth and our revenue, that is the primary goal, that is the structure of the way that the game and society has been set up right now. Now, obviously, there are rules on what, for example, VCs, we have been told not to invest in companies, or help from a certain country, or countries, there’s, there’s limits, that those countries are not aligned with the Western goals and values. And so, that is a clear law and regulation. So, to answer your question, very, very point blank. The goal is on business growth within the laws, and they do need to align to humanity, otherwise, they just won’t sustain.
Minter Dial 48:47
And that’s for the pragmatic reason that they won’t actually exist continue to exist.
Jeremiah Owyang 48:53
That’s right. And therefore, they could not complete their mission that
Minter Dial 48:57
fabulous, Jeremiah, I expected nothing less than a highly stimulating conversation with you. Thank you. wonderfully fun to have this. It’s so it’s extremely exciting for me and I have these little goose bumps going up but to be able to talk to somebody who’s so within this, this part of the industry and I think that you are a beacon for us all you certainly have helped me I got lots of fun things to add to the show notes. But how can people follow you follow your writings Jeremiah, what’s the best way to get in touch or at least you know, see what you’re up to and follow your words.
Jeremiah Owyang 49:31
Oh, thanks Minter. So, I am on social and my handle is J and my last name. So, Owyang which is O-W-Yang. And that’s very easy to find or just Google Jeremiah Owyang and find me on socials and newsletter blog. It’s all out there. Thank you for having me, Minter.
Minter Dial 49:48
In my great pleasure. Stay in touch Jeremiah. Let’s hopefully get one of the chances to your Llama Lounge or somewhere else hug in an old-fashioned way.
Jeremiah Owyang 49:56
I would look forward to seeing you thank you for your time.
Minter Dial 50:03
So a really heartfelt thanks for listening to this episode of The Minter Dialogue podcast. If you liked the show, please remember to subscribe on your favourite podcast service. As ever, rating and reviews are the real currency of podcasts. And if you’re really inspired, I’m accepting donations on www.patreon.com/Minterdial. You’ll find the show notes with over 2100 blog posts on minterdial.com on topics ranging from leadership to branding, tech and marketing tips. Check out my documentary film and books including my last one, the second edition of “Heartificial Empathy, Putting Heart into Business and Artificial Intelligence” that came out in April 2023. And to finish here’s a song I wrote with Stephanie Singer, “A Convinced Man.”
I like the feel of a stranger
Tucked around me
Precipitating the danger
To feel free
Trust is the reason
Still I won’t toe the line.
I sit here passively
Hope for your respect
Anticipating the thrill of your intellect
Maybe I tell myself
There’s no use in me lying.
I’m a convinced man,
Building an urge
A convinced man,
To live and die submerged.
A convinced man,
In the arms of a woman
I’m a convinced man
Challenge my fate
I’m a convinced man
A convinced man
In the arms of a woman.
And struggle to see
Live for the challenge
So life’s not incomplete
What’s wrong with challenge
I know soon we all die
I’m a convinced man
Practicing my lines
I’m a convinced man
Here in these confines
A convinced man
In the arms of a woman.
I’m a convinced man
Put me to the test
I’m a convinced man
I’m ready for an arrest
I’m a convinced man
In the arms of a woman.
I’m a convinced man… so convinced
You convince me, yeah baby,
I’m a convinced man
In the arms of a woman…
Minter Dial is an international professional speaker, author & consultant on Leadership, Branding and Transformation. After a successful international career at L’Oréal, Minter Dial returned to his entrepreneurial roots and has spent the last twelve years helping senior management teams and Boards to adapt to the new exigencies of the digitally enhanced marketplace. He has worked with world-class organisations to help activate their brand strategies, and figure out how best to integrate new technologies, digital tools, devices and platforms. Above all, Minter works to catalyse a change in mindset and dial up transformation. Minter received his BA in Trilingual Literature from Yale University (1987) and gained his MBA at INSEAD, Fontainebleau (1993). He’s author of four award-winning books, including Heartificial Empathy, Putting Heart into Business and Artificial Intelligence (2nd edition) (2023); You Lead, How Being Yourself Makes You A Better Leader (Kogan Page 2021); co-author of Futureproof, How To Get Your Business Ready For The Next Disruption (Pearson 2017); and author of The Last Ring Home (Myndset Press 2016), a book and documentary film, both of which have won awards and critical acclaim.
👉🏼 It’s easy to inquire about booking Minter Dial here.