Latest / General Magic & The Soul of Tech / Ep. 12 | “Schrambo” Ships: AI, Climate Tech, and Advice for Founders | Steve Schramm

Transcript

Ep. 12 | “Schrambo” Ships: AI, Climate Tech, and Advice for Founders | Steve Schramm · General Magic & The Soul of Tech

0:00
SPEAKER_0: Hi, this is Mike Stern.
0:03
SPEAKER_1: And this is Tom "Hersh" Hershenson.
0:05
SPEAKER_0: Welcome to General Magic & The Soul of Tech.
0:10
SPEAKER_2: We believed we were building something that would change the world, and in some ways we did.
0:14
SPEAKER_2: In others, we didn't understand what we were setting in motion.
0:18
SPEAKER_2: The story of General Magic isn't over.
0:21
SPEAKER_2: We're still living inside what it started.
0:29
Our special guest today is Steve Schramm,
0:32
one of the OG General Magicians and a real Silicon Valley veteran.
0:38
SPEAKER_0: I'm gonna bring back something from the past.
0:40
SPEAKER_0: Steve's nickname at General Magic when he was VP of Engineering was Schrambo.
0:45
And the reason he had that nickname was because of his hard driving,
0:49
we gotta ship it, we gotta stop fooling around style.
0:54
SPEAKER_0: That name really suited you.
0:56
SPEAKER_0: And you were a real manager, which was something we were rather short of.
1:01
And my initial question is,
1:03
you came up through big companies,
1:05
HP and McDonnell Douglas, as well as startups before you got to Magic.
1:10
SPEAKER_0: And you weren't part of the Apple clan, which was part of the General Magic DNA.
1:15
SPEAKER_0: Did that make a difference?
1:16
Was that one of the reasons why you were,
1:18
much more focused on meeting deadlines and executing than the rest of us?
1:22
SPEAKER_3: Well, yes, I think it's pretty clear that my background made a difference.
1:27
SPEAKER_3: Where I had been previously, it was all about delivering.
1:31
SPEAKER_3: There was no artistry in it.
1:33
SPEAKER_3: And you can view that as a failing on the corporate side, but it really was about delivering.
1:38
The other thing about it was that not only did I not come from Apple,
1:42
but I really didn't come from the hardware and software technology side of it.
1:47
SPEAKER_3: I really came from the communication side.
1:49
And working with Jim White and Rich Miller and other folks in
1:54
the International Telecommunications Union in the United Nations,
1:58
where we were setting worldwide data communication standards,
2:02
it was all about reaching agreement and getting things out.
2:05
SPEAKER_3: And you had a deadline for doing that.
2:08
SPEAKER_3: And you worked worldwide with a lot of very different people with different perspectives.
2:12
And so You're constantly in the middle of trying
2:15
to understand everybody's perspective,
2:17
but saying, look, we have to deliver a standard and we have to agree.
2:21
SPEAKER_3: And so for me, Jim White and Rich Miller were the rock stars.
2:24
SPEAKER_3: And while I used Macintosh devices, I lived the communication side and we had to deliver stuff.
2:30
So yeah,
2:31
it's a different perspective than the more software and change the world innovation
2:36
that Apple had.
2:37
SPEAKER_0: Yeah.
2:38
SPEAKER_0: And I think, the we don't need any managers ethic was foreign to you.
2:44
SPEAKER_3: Yeah.
2:45
SPEAKER_3: The managers were supposed to provide value.
2:47
We were supposed to keep our eye on the ball and look at the big picture and
2:51
then figure out how that applies to getting things done.
2:54
Obviously to get things done,
2:56
whether it's in telecommunications or whether it's in software and hardware,
3:00
all sorts of problems come up and you've got to understand them and deal with them.
3:03
SPEAKER_3: And sure, things sometimes take longer than you expect.
3:06
But the trick is to Try to bite off the right amount and then when things go wrong,
3:12
figure out how to get them right again.
3:14
SPEAKER_0: Yeah, you're wearing your AT&T t-shirt.
3:16
SPEAKER_0: You should show people the back.
3:18
SPEAKER_3: Yeah.
3:19
SPEAKER_3: Yeah.
3:19
SPEAKER_3: These are what we called fuzzy balls of intelligence.
3:23
SPEAKER_3: Yeah, this is AT&T.
3:24
SPEAKER_3: This is 1990s, right?
3:26
Early 1990s,
3:27
and they were gonna have agents running around in their network doing bidding
3:33
for all your personal and professional and corporate needs.
3:37
SPEAKER_3: These were customized agents.
3:39
SPEAKER_3: We had worked really closely with AT&T Bell Labs.
3:43
SPEAKER_3: They had thought through really carefully how do you protect the network against these agents?
3:48
SPEAKER_3: How do the agents have the features and capabilities to deliver services to their customers?
3:54
How do they have access so that AT&T can interrogate them,
3:58
manage them,
3:58
make sure they're not consuming too many resources, make sure they're not abused.
4:03
We went through a lot of work with some really smart people,
4:08
over years,
4:09
to make sure that we designed Telescript,
4:11
which was the language that you programmed the agents in,
4:14
to make sure that it was manageable and maintainable and supportable.
4:18
SPEAKER_3: It delivered the goods, but also didn't trash the network.
4:21
SPEAKER_3: And AT&T was proud of the work that they did in this area.
4:25
SPEAKER_3: It was a new AT&T that was turned loose and really wanted to impact the world.
4:30
And they were proud of the work on Telescript,
4:32
on the language, and they wanted to share it with the other PTTs around the world.
4:36
SPEAKER_3: And some of the PTTs that understood what Telescript was doing, they embraced this.
4:41
SPEAKER_3: They said, wow, yeah, smart network.
4:43
SPEAKER_3: This makes sense.
4:44
We could provide all sorts of services to our customers and we could also manage it
4:49
and take care of it.
4:50
I know NTT and Cable & Wireless and other PTTs around the world really embraced this
4:57
and wanted to take off with it.
4:58
So It was an exciting time and it was really,
5:01
I think,
5:01
the first time that the top folks around the world thought about how do you make
5:06
a network intelligent?
5:07
SPEAKER_3: How do you give it the ability to deliver products and services?
5:11
So we were on the forefront of that and now we're seeing a second or maybe even
5:15
a third wave of that happening.
5:17
I'm curious with respect to the design and the concerns that the PTTs had,
5:22
were those driven by a purely technical perspective?
5:26
SPEAKER_1: Or was there some thought about the possible misuse of agents and malicious actors?
5:32
SPEAKER_1: What drove embedding the kind of controls and safety that Telescript had?
5:39
SPEAKER_3: Yeah, Hersh, it was totally the latter.
5:40
They have experience,
5:42
decades of experience of people trying to abuse the system,
5:46
both maliciously and unintentionally.
5:48
They have to make sure that their circuitry,
5:52
and of course,
5:52
back then it was mostly copper wire,
5:55
they had to make sure that the wire wasn't abused,
5:58
that it could handle all of the all the things that were happening.
6:01
SPEAKER_3: And that also goes for non-malicious things.
6:04
You have an earthquake,
6:05
everybody wants to call and check and make sure people are okay
6:08
and people are calling out trying to get help.
6:11
SPEAKER_3: So they really thought through those kinds of things.
6:14
You can think of that as a denial of service attack,
6:16
but actually it's just the people using the network.
6:19
So they were all about malicious use because they said,
6:22
whoa, wait a minute, you're talking about programming our network.
6:26
SPEAKER_3: We want to make sure that people can only program it for good and not for evil.
6:30
And when they try to program it for evil,
6:32
because we know they will,
6:34
we want to be able to detect it and basically box it so that it can't get out
6:38
and do evil stuff.
6:40
SPEAKER_3: And so, yeah, absolutely.
6:41
SPEAKER_3: We had concepts of agents and places in the network.
6:45
And so an agent would travel to a place,
6:47
but the place could say, okay,
6:49
I'm giving you only this many resources,
6:51
and it monitors that and the place can report back saying, hey,
6:56
this agent is trying to abuse its resource consumption and I'm gonna shut it down
7:00
or control it.
7:01
SPEAKER_3: That was just one example.
7:02
There were many different ways in which agents were tracked,
7:05
monitored,
7:06
and metered so that the network would operate well with billions
7:11
of these agents running around.
7:13
SPEAKER_3: So yeah, the Bell Labs guys were all about it.
7:16
Yeah, we were thinking of cracking the alliance very late in the game
7:20
before we went public and bringing in of all things, IBM.
7:24
SPEAKER_0: And that was when you formulated what became known as the Schramm Paradox.
7:29
Can you explain to our listeners what that was and how that insight carried forward
7:35
to other places where you worked and even today.
7:39
SPEAKER_3: Yeah, it was an interesting situation.
7:41
SPEAKER_3: We, we're trying to invent an entire world, right?
7:45
Not just the iPhone,
7:47
but the iPhone and the internet and everything that you have today
7:50
through cellular networks.
7:52
SPEAKER_3: We're trying to invent all of that at once.
7:53
SPEAKER_3: But at the same time, we were a business and we needed to make money.
7:56
SPEAKER_3: And so we were trying to figure out how do we charge for all of this?
8:01
And so you get into these real strange things,
8:05
which and I started to describe some of these when I was talking with Mike
8:09
and others about how we price stuff.
8:11
We got into these weird situations where,
8:14
for example,
8:15
on the AT&T side, we wanna see lots of agents, but we also need to charge.
8:20
SPEAKER_3: Well, okay, one idea would be, well, let's just have a very small charge for each agent.
8:26
SPEAKER_3: Well, if you think about that, at a superficial level, you go, Yeah, sure, whatever.
8:30
SPEAKER_3: Okay, go ahead.
8:31
SPEAKER_3: At a deeper level, you're like, well, wait a minute.
8:34
If you're charging for each agent,
8:36
then you're effectively discouraging people from using agents
8:39
because every agent they use costs money.
8:42
And while that may not seem like very much,
8:44
for one,
8:45
if you do a million or a billion, the numbers start to get very, very large.
8:49
And so what you end up having is this tension where on the one hand,
8:54
we want to make a lot of money, so we want to charge for something.
8:57
And yet the more of something that we charge for,
8:59
the more our customers are motivated to not use them.
9:03
SPEAKER_3: And even if you try to, provide volume discounts and stuff like this, it gets very, very sticky.
9:09
SPEAKER_3: You've got to do the calculation.
9:10
And so that's what Mike's referring to when we're talking about
9:14
the Schramm Paradox is like you can't just extract money.
9:17
You have to figure out what motivations you're creating or creating disincentives
9:23
for when you do these setups.
9:25
SPEAKER_3: And that repeats itself with every company that I'm working on, right?
9:29
SPEAKER_3: Because I do startups.
9:30
SPEAKER_3: That's what I do.
9:31
I'm advising about a dozen right now,
9:33
mostly in climate change mitigation and clean energy and medical breakthroughs
9:37
and stuff.
9:38
SPEAKER_3: And every one of them needs money.
9:40
And so, yes,
9:41
you go through venture capital and strategic investors and other things like that
9:44
to get going,
9:45
but you want to stop doing that and start making money off of your business.
9:49
You start charging for stuff,
9:51
but you've got to be really careful because if you're charging for things
9:54
that your customers are now motivated to not use,
9:56
then your business model doesn't work.
9:58
And so it's really important to charge for value and increase the value so
10:04
that people feel like,
10:06
Yes, I'm spending money, but I'm getting enough value for it to justify it.
10:10
SPEAKER_0: So the frontier models are confronted with this right now, right?
10:13
SPEAKER_0: In terms of charging per token or how they price, right?
10:16
I just want to have a quick digression to that exact point
10:19
because I'm heavily using Anthropic stuff.
10:24
I'm also using Chat and Gemini and other things,
10:26
but I have written skills for Anthropic to do my bidding so that I can manage all
10:32
the work that I'm doing.
10:33
SPEAKER_3: And I wrote skills, they were working great.
10:37
SPEAKER_3: I was using them constantly, and then they changed the billing model.
10:41
SPEAKER_3: And the skills that I created are using the API.
10:45
SPEAKER_3: They've decided to charge that differently, and I blew through my limit in two days.
10:50
SPEAKER_3: And so I looked at that and I said, Okay, I guess you don't want me to use your API anymore.
10:56
SPEAKER_3: Fine.
10:57
SPEAKER_3: And I switched all of my work.
10:58
SPEAKER_3: I restructured them and I used some Anthropic and some Chat stuff now.
11:02
SPEAKER_3: But I'm not doing that anymore.
11:04
SPEAKER_3: I just killed the skill and I'm just not using it.
11:06
SPEAKER_3: I found other ways to get it done.
11:08
SPEAKER_3: So they motivated me to stop using their product and service.
11:13
SPEAKER_3: And you know, oops.
11:15
SPEAKER_3: So that's the Schramm Paradox back at you again.
11:20
SPEAKER_0: Yeah, what's a skill?
11:21
SPEAKER_0: What's a skill, Steve?
11:25
SPEAKER_3: In Anthropic's world, you can effectively program their model to do stuff for you.
11:36
SPEAKER_3: In English, you explain what you want it to do, and it takes that in and it operationalizes it.
11:45
SPEAKER_3: It's effectively a high-level way of programming, but you're not writing any language.
11:49
The whole, the,
11:51
I mean this – this whole AI thing is just so amazing because I can do in one day,
11:59
I could do what would have taken me weeks, maybe months to do.
12:03
And I'm moving so fast now that it,
12:06
it's enabling me to do things that I never imagined that I could do.
12:11
So you can create this skill,
12:13
and once you've created it then it just does it all the time for you.
12:17
SPEAKER_3: You don't ever have to do it again.
12:18
So just to give you a specific example,
12:21
I wrote a skill that enabled,
12:23
and this will scare you,
12:24
anybody with privacy concerns,
12:27
I urge you to turn off your volume at this moment because you don't want
12:30
to hear this.
12:30
SPEAKER_3: But I I've embraced this.
12:33
I understand the privacy concerns,
12:35
but what I did was I programmed this so that it will go and look at all of my email.
12:41
There are connectors for my phone,
12:44
it will pull things,
12:45
it can talk to my to-do list,
12:47
it can go to my calendar,
12:49
it looks at everything that I'm doing,
12:50
it looks at all the email that I'm doing,
12:52
it keeps track of all the companies that I'm working on,
12:55
and it stores all this data locally.
12:56
This is part of the skill that I wrote for it is it stores all the data
12:59
on my machine here so it doesn't go into the cloud.
13:03
And so then what happens is when I ask it a question,
13:07
it can go and look at the data locally and say,
13:09
okay, I understand these last eight years you've been working with this company.
13:13
SPEAKER_3: I understand all the people.
13:14
SPEAKER_3: I understand the context.
13:16
And so when I ask it a question, hey,
13:19
this is the conflict that they're having or this is the go
13:22
to market challenge they have or whatever,
13:24
it has all that context and it can then have a discussion with me about Well,
13:29
they could try this or this, or have you thought of that?
13:31
SPEAKER_3: Or let me go out and research this for you.
13:32
SPEAKER_3: And I'll engage in a super quick, very efficient conversation with it to decide what to do.
13:37
And then I'll say, okay, great,
13:39
put that on my to-do list or throw that on my calendar and create this or that,
13:43
and it will do it for me.
13:44
SPEAKER_3: And I can move on to the next thing.
13:46
We've been exploring with Magicians that we've been talking to how they're using AI,
13:51
And, we've been very reluctant to use it for fear of rotting our brains, as writers.
13:57
SPEAKER_0: And what we're hearing from you and others is it doesn't rot your brain.
14:02
SPEAKER_0: It empowers it.
14:03
SPEAKER_0: And that's a significant reset for us and for me.
14:07
SPEAKER_0: And that's a sterling example, what you've just described.
14:10
SPEAKER_0: How is it propagating inside the portfolio companies you're working with?
14:14
SPEAKER_0: And what are doing to teach people how to use it well and avoid missteps?
14:19
Yeah, so since I have been talking fairly enthusiastically about it,
14:23
I do wanna caution everybody that it's not perfect.
14:26
I know you know that,
14:27
but it's imperfect in major and significant ways that you really have to watch.
14:33
SPEAKER_3: People used to talk about hallucinations.
14:35
It doesn't really do that anymore,
14:37
but sometimes it will State something which sounds like it's real, but it's not.
14:42
SPEAKER_3: It's just not correct.
14:44
It may be completely incorrect,
14:46
or it may be that it's asserting something that you need to check.
14:50
So it's very much like talking to a real person in the sense that
14:54
a real person can tell you things.
14:55
SPEAKER_3: I could be telling you things, and you shouldn't just believe what I'm saying.
14:58
SPEAKER_3: You should go figure it out for yourself.
15:00
And when you do figure it out for yourself,
15:02
you may not come to the same conclusions that I came to.
15:05
SPEAKER_3: So it's very much like that.
15:07
It's been trained on data that's been created by people and the behaviors are
15:11
from people and the conflicts and the disagreements are all from people.
15:15
SPEAKER_3: So in some senses it has some of those same attributes.
15:19
SPEAKER_3: And one of them is it's not right.
15:21
SPEAKER_3: And so when it tells me things or when we're having a conversation, let me just be specific.
15:28
I'm having a conversation with it about a strategy from one of my companies,
15:32
it will say something and I'll say,
15:36
wait a minute, are you basing that on something that the company previously did?
15:40
SPEAKER_3: And it'll say, yeah, it's the it did this before.
15:43
SPEAKER_3: And I said, when?
15:44
SPEAKER_3: And they'll say, well, okay, it wasn't exactly that.
15:47
SPEAKER_3: It was this other thing.
15:48
SPEAKER_3: So you have to call it on these things.
15:50
SPEAKER_3: You can't just accept what it says.
15:52
SPEAKER_3: You cannot just sit there and go, it's coming from a computer, therefore I should go with it.
15:58
SPEAKER_3: And just continue with it.
15:59
SPEAKER_3: You have to challenge it.
16:01
SPEAKER_3: And this is true in real life too.
16:03
You don't believe everything politicians say,
16:05
you don't believe everything that leaders say, any hierarchy.
16:09
SPEAKER_3: Even the church hierarchy has issues.
16:11
SPEAKER_3: So it's not anything new.
16:13
And I guess that's what I would encourage everybody to do is just to say,
16:16
well, okay, it's just another source of information.
16:18
SPEAKER_3: It's a tool.
16:20
SPEAKER_3: Use it, but use it within the bounds that you need to keep around it.
16:25
Yeah, we spoke to Steve Jarrett and one of the things that he talked about was AI
16:31
to challenge things,
16:32
a kind of devil's advocate,
16:33
and he would pit different AI models against each other, kicking the tires on ideas.
16:38
SPEAKER_1: So not something to make decisions with, but something to validate or invalidate hypotheses.
16:44
SPEAKER_3: Yeah.
16:45
SPEAKER_3: It's great for that.
16:46
SPEAKER_3: And I do that all the time.
16:47
In fact,
16:47
before we got on this call,
16:49
I was working on a problem between two founders in a startup that I have.
16:54
SPEAKER_3: And I actually use Claude and Chat.
16:57
I will feed them the same information,
16:59
get different answers from them, and then have them argue between themselves.
17:04
And the process of watching them either converge or diverge is really helpful for me
17:10
because then I can take all of the best things and ignore all the BS
17:15
and provide some useful feedback.
17:17
SPEAKER_3: So absolutely I do that.
17:20
For entrepreneurs in early stage companies,
17:23
it's just a godsend because one of the big reasons people don't start companies is
17:28
because they don't do a lot of the basic stuff well.
17:30
SPEAKER_3: They don't create a pitch for themselves that is effective.
17:34
SPEAKER_3: They don't come up with a business plan that's effective.
17:36
Well, AI can help with these things,
17:39
and this is enabling smart people who have good ideas to actually bring them forward
17:44
in a coherent manner so that People who are potential investors
17:48
or supporters can help them get off the ground.
17:52
SPEAKER_3: A lot of them just don't get off the ground at all.
17:54
SPEAKER_3: And right now in our economy, there's so much money looking for good investment.
18:00
And if you can present a coherent idea,
18:02
even if you're not a great business person,
18:04
you can get in touch with somebody who says, I understand what you're trying to do.
18:08
SPEAKER_3: Let me partner you up with somebody who can really help you run this business.
18:12
SPEAKER_3: And then you get a couple of people together.
18:15
SPEAKER_3: Somebody with a great idea, somebody with a great business sense.
18:18
And then they can use AI to put together their company really fast
18:22
and get some investment and go.
18:24
And so I see AI as a really powerful tool to get more good ideas
18:29
and more good things out into the marketplace.
18:33
SPEAKER_3: And I think that can improve our life in a lot of different ways.
18:36
SPEAKER_3: And I'm seeing it used aggressively by startups.
18:39
At Stanford we have StartX – We take a cohort of about 40 companies every quarter
18:45
and we school them up on how to run a business.
18:48
SPEAKER_3: And we're seeing really high quality ideas come into that.
18:51
SPEAKER_3: And it's just it's great to see.
18:53
You pivoted from communications and device technology to climate tech and
18:59
to climate change mitigation.
19:01
Can you talk about the portfolio companies that you're working with a little
19:05
and why you ended up in that sector?
19:07
SPEAKER_3: Yeah, I ended up in the sector because I wanted to do something useful.
19:12
SPEAKER_3: I'd done a bunch of startups over the decades and most of them were tech.
19:18
SPEAKER_3: I wanted to focus on something that was trying to make the world better.
19:23
And I settled on climate change mitigation as an area where there were a bunch
19:28
of different ideas.
19:29
SPEAKER_3: I didn't know which one would work.
19:30
And so I figured,
19:31
well, if I could help a bunch of them at once, maybe could see some good results.
19:35
SPEAKER_3: And so, I just started picking them up, right?
19:37
Every time a cohort of companies would come in,
19:42
I would look through them and look for ones that were doing something
19:45
that made sense to me that I thought I could help with.
19:48
SPEAKER_3: It also gave me an excuse to learn what's going on in the technology.
19:53
So for example,
19:54
the first one I picked,
19:55
10 years ago,
19:56
a company called SkyCool,
19:59
they were taking advantage of a physical property of light where there's
20:03
a certain range,
20:05
it's just in the infrared,
20:06
where if you move heat, is just electromagnetic radiation from the sun.
20:11
SPEAKER_3: If you shift it, if you shift the wavelength, it actually goes out into space.
20:15
SPEAKER_3: It doesn't get trapped by the atmosphere.
20:17
SPEAKER_3: And so their idea was we'll create this thin layer of film.
20:21
And what it'll do,
20:22
it'll have 130 layers of plastic,
20:25
and it'll move the sun's wavelengths to that wavelength where it can escape
20:30
into space.
20:31
SPEAKER_3: And that way, if you put this on something, then it will make that object cool.
20:36
SPEAKER_3: Even if it's hot out.
20:37
SPEAKER_3: That was the idea.
20:38
SPEAKER_3: It turns out it works remarkably well.
20:41
SPEAKER_3: You can put this film, they eventually contracted with 3M to have 3M manufacture it.
20:47
And if you put this on an object,
20:49
now obviously it depends on relative humidity in the area and stuff,
20:52
but it can lower the surface temperature by 10 to 15 degrees Fahrenheit
20:58
over what it is.
20:59
SPEAKER_3: So it's a remarkable amount.
21:01
SPEAKER_3: And so the company's doing it in two different ways.
21:04
SPEAKER_3: The obvious way is fine.
21:05
You could just lay it on top of a mobile home or lay it on top of a delivery vehicle
21:09
and it brings the temperature down.
21:11
SPEAKER_3: That's nice.
21:12
But what they're also doing is they're going to grocery stores
21:16
who have these giant cooling sections.
21:18
SPEAKER_3: So they're chewing huge amounts of energy to keep things cool.
21:21
And what you do is you run water through panels,
21:24
they look like solar panels, but instead the panels have the film on it.
21:27
SPEAKER_3: And so the water is cooled by 10 to 15 degrees.
21:30
SPEAKER_3: And that means the stores need less power to keep all of the refrigeration or freezing cold.
21:37
SPEAKER_3: It's a dramatic reduction in energy use.
21:40
SPEAKER_3: And so that was the first thing.
21:42
SPEAKER_3: And for me as a as an engineer and a scientist, I was like, wow, that's really amazing.
21:47
SPEAKER_3: Let's figure out how to make the company go.
21:49
SPEAKER_3: And so we've been working on that.
21:51
SPEAKER_3: So that was the my introduction to it, and I got very excited about it.
21:54
SPEAKER_3: I could rattle off a bunch of other really cool ideas.
21:57
One is this idea of wow,
21:59
there are 2 million oil well holes in the United States and Canada that are dry.
22:04
SPEAKER_3: They're dead, they need to be cleaned up.
22:06
SPEAKER_3: And so they're useless and actually a liability on the balance sheet of the oil companies.
22:10
SPEAKER_3: Meanwhile, We have solar power and wind power, which is sporadic.
22:16
Sometimes there's a lot of it,
22:17
too much of it, sometimes even, and sometimes there's none or not enough.
22:21
SPEAKER_3: Why don't we combine those two problems together and fix them?
22:24
We will take an old oil well,
22:27
we will clean it up,
22:29
and then we will put a very large weight on a winch at the top of the well.
22:35
SPEAKER_3: We'll connect that up to the grid.
22:37
And then when the grid needs power in the middle of the night when it's not windy,
22:41
you drop the weight and it falls a mile and spins and provides power.
22:46
And then when you have an excess of solar power or an excess of wind power
22:51
and it's very,
22:51
very cheap or almost free or even negative cost,
22:54
you use that power to bring the weight back up.
22:57
I thought this was a phenomenal idea because it cleans up a liability,
23:01
a danger to the environment,
23:03
and it creates a kinetic battery which provides you power
23:07
when your inconsistent clean power is not available.
23:11
SPEAKER_3: Seemed like just a great idea.
23:13
SPEAKER_3: And so we, came up with a prototype.
23:15
SPEAKER_3: The prototype works.
23:16
SPEAKER_3: Out in Bakersfield we created a real one.
23:19
SPEAKER_3: It it's our alpha, it works.
23:21
And so now they're signing contracts with oil companies who seem willing
23:25
to get these liabilities off their books and stuff.
23:28
SPEAKER_3: I just thought that was a really great idea.
23:29
SPEAKER_3: So that's very cool.
23:31
SPEAKER_3: And we have another company that methane.
23:34
SPEAKER_3: Methane is a very powerful greenhouse gas.
23:37
SPEAKER_3: It's really bad.
23:38
SPEAKER_3: It's hundreds of times worse than carbon dioxide.
23:41
And there is a lot of methane leaking from all sorts of natural sources as well as
23:47
from man made sources.
23:49
And so this company Said, okay, well,
23:53
for methane that's coming out of an oil well,
23:55
if you just want to burn it, you can burn it.
23:57
Because there's a steady flow,
23:58
a constant flow of methane,
24:00
and so it combusts and turns into carbon dioxide and other things.
24:04
SPEAKER_3: And that's not great, but it's not as bad as letting the methane go up.
24:08
SPEAKER_3: But here's the thing, In other sources, you don't get a steady flow of methane.
24:13
For example,
24:13
just something you didn't think about probably,
24:15
but there are a lot of coal mines in the world,
24:18
and coal mines emit huge amounts of methane,
24:21
but they don't do it in a consistent manner.
24:23
So what you need is something to combust the methane in a way where it will continue
24:28
the combustion to basically turn that methane into carbon dioxide,
24:33
which is much less dangerous so this person she invented this device
24:39
which actually does that and then created a company around it.
24:43
And so now we're going around to places that generate methane
24:46
in inconsistent manners, such as coal mines, but also such as dairy cattle.
24:51
SPEAKER_0: Ha.
24:51
SPEAKER_3: They.
24:52
SPEAKER_0: The Ronald.
24:52
SPEAKER_3: They...
24:53
SPEAKER_0: Reagan problem, cow farts, right?
24:55
SPEAKER_3: That's right.
24:56
SPEAKER_3: It farts and poops, right?
24:58
SPEAKER_3: So you've got giant pools of stuff that are emitting methane.
25:03
SPEAKER_3: So you put a little roof over that and then you can get rid of the methane that way.
25:08
SPEAKER_3: I have two companies working on green ammonia.
25:10
Now you may not realize this,
25:12
but ammonia is critical for the entire world in terms of producing food.
25:17
SPEAKER_3: That's how you move nitrogen around, right?
25:19
Ammonia is an NH3,
25:20
you've got nitrogen,
25:22
but you get a bonus because it's really hard in general to move hydrogen around.
25:26
SPEAKER_3: Hydrogen are really little, teeny little atoms, and they tend to go through everything.
25:31
SPEAKER_3: It's very hard to contain them.
25:32
SPEAKER_3: Even if you put them in a pipeline, it makes the pipeline very brittle and it breaks over time.
25:37
SPEAKER_3: So ammonia is a great way to move hydrogen and nitrogen around.
25:41
SPEAKER_3: Hydrogen being a fuel, very, very nice fuel, and nitrogen being a great fertilizer.
25:47
SPEAKER_3: So there is a huge infrastructure in the world to move ammonia around.
25:51
The problem is ammonia is something that is hard to produce,
25:55
and the current market way to make it is this process that requires a huge amount
26:01
of energy and pressure.
26:02
SPEAKER_3: And that's usually used with fossil fuels.
26:05
So I have two different companies that have come up with ways
26:08
to create what we call green ammonia or ammonia that's produced without fossil fuels
26:14
and very clever ideas to do it
26:15
and they've got some really good scientific breakthroughs on this
26:19
which I think have legs so we we're getting funding for that and moving
26:22
that forward.
26:23
They're all using AI as well And I have another company that is working on
26:29
the greenhouse gas side, for refrigerants.
26:32
SPEAKER_3: There are really yucky chemicals that are used for that.
26:35
Everybody knows chlorofluorocarbons,
26:37
but there are other things even that have replaced them that are really bad for
26:41
the environment.
26:42
This guy has come up with a way to use carbon dioxide for data centers,
26:46
because data centers right now are just generating massive amounts of heat and
26:52
the air cooling doesn't really work anymore and water cooling is running
26:57
into its maximal spot and he's come up with a way to use carbon dioxide as a coolant
27:01
so moving that forward.
27:02
Anyway there are other companies I'm working with but the idea is that there are
27:06
a lot of very smart people with a lot of energy just like
27:10
at General Magic right it's just right in that peak area where they know enough
27:14
to be valuable and they've got energy And they are charging ahead
27:18
and doing all sorts of great stuff.
27:21
And for me,
27:22
it's very exciting to be working with them and I'm learning just a ton of stuff
27:27
about how to do this.
27:28
Because while AI is helpful,
27:31
you still need real people to come up with these innovations and bring them
27:35
to market and get them into the marketplace.
27:37
SPEAKER_3: So that's how I'm spending my time right now.
27:40
SPEAKER_0: Yeah.
27:40
SPEAKER_0: Are you getting market uptake?
27:42
SPEAKER_0: Sounds like they're getting funding, but are they getting customers?
27:45
SPEAKER_3: Yeah, some of these are early stage.
27:47
SPEAKER_3: And so the answer is uncertain.
27:50
SPEAKER_3: But some of the ones that I've been working with for a while are getting market uptake.
27:55
One of them,
27:55
the one I've been working with the very longest,
27:59
is a company that started out doing automated farm management software for cattle.
28:03
And you before you say, Well,
28:04
what does that have to do with climate change,
28:07
there are over a billion cows on the planet.
28:09
SPEAKER_3: And if you weighed all the cows, they would weigh more than all the people on Earth.
28:14
SPEAKER_3: So it's a lot of mass.
28:16
SPEAKER_3: And it's very important, obviously, the food supply.
28:19
So if you come up with ways to manage the all those animals,
28:24
both cattle and sheep,
28:25
actually,
28:25
there are a lot of sheep too, then you can reduce the impact on the environment.
28:31
You can cut your carbon costs,
28:32
you can reduce the impact that they're having on the air and on the water and
28:37
the land.
28:37
So that company started from nothing and has built up a product that is now
28:43
in use around the world in many different countries.
28:46
They have thousands of farms,
28:48
many thousands of farms,
28:49
probably over 10,000 now,
28:51
and I think over 25 to 30 million animals under management.
28:56
So a lot of animals are being tracked and managed and their carbon usage reduced
29:03
but more importantly all that data is providing us with input into our own models
29:08
to figure out how to even be more effective.
29:10
And this is something where we now have consortiums,
29:14
as you might expect,
29:15
the largest companies in the world, Nestlé,
29:17
who does pet food,
29:19
McDonald's,
29:20
who does burgers, RBI,
29:21
who does Burger King,
29:23
all those people are now working with us,
29:25
with this company,
29:27
in an attempt to actually reduce carbon usage through carbon insetting
29:31
within their corporate hierarchies and within the supply chain.
29:34
So it's a really exciting thing to see because it took many years to get
29:39
to this point,
29:40
but we are seeing successful use of AI and data to reduce carbon impacts.
29:45
So you you've experienced patient money and people who have worked for a decade
29:50
or two decades and have finally had market success.
29:54
SPEAKER_0: Why is that possible?
29:57
SPEAKER_0: To persist that long and end up eventually get there?
30:01
SPEAKER_0: Where's the difference?
30:04
SPEAKER_3: It's interesting.
30:05
SPEAKER_3: I don't know that there's one answer to that, but I can tell you some contributing factors.
30:10
One of them is you have to have people who are committed to the solution
30:16
and building a successful business,
30:18
but also really respect each other, have a lot of integrity and work well together.
30:22
SPEAKER_3: That's not the answer, but it's table stakes.
30:25
Because you can't work together for a very long time and make all the changes
30:29
and stay with something if you don't have that.
30:31
SPEAKER_3: And I know we had that at General Magic, but it takes more than that.
30:35
SPEAKER_3: The second thing is you need to have the right kind of investors.
30:39
SPEAKER_3: I was telling you about the cattle company.
30:42
They had investors that were very very,
30:44
very patient that understood that they were entering into
30:48
a very long-term relationship that would be decades.
30:51
SPEAKER_3: That was their expectation.
30:53
They weren't expecting a quick win,
30:55
they weren't expecting a unicorn company,
30:58
and they weren't expecting dramatic growth.
31:00
So if you have those kinds of investors,
31:03
then your board meetings are a lot different than some of the ones
31:06
that we've experienced in other places.
31:08
SPEAKER_3: Because they're not expecting every quarter to be big jumps and growth and things.
31:12
SPEAKER_3: They want to see a slow, steady improvement in the company and its traction.
31:19
I think another thing is a willingness to change and pivot when you need to
31:24
and doing that at the right time.
31:25
SPEAKER_3: In retrospect it's easy to do, but at the moment it's hard.
31:30
And that's something that I've seen work with some of the companies
31:32
that I've been working with for 10 years now.
31:35
SPEAKER_3: So it's definitely a combination of multiple things.
31:40
I believe there is room to create value,
31:42
to create something that's not high growth,
31:45
but instead provide solid continuous value to people.
31:49
And so when I see these businesses,
31:52
I frequently tell them don't necessarily go after venture capital.
31:56
Think about if you're trying to do something that's good,
31:59
find the right types of investors for your company.
32:02
SPEAKER_3: I'm I was just talking to a company.
32:04
They have a breakthrough technology for detecting forest fire ignition
32:09
before it can even be seen visually.
32:12
SPEAKER_3: And it's great.
32:13
SPEAKER_3: I think it's fantastic and they're deploying it now and testing it out.
32:17
But he was asking me about financing and,
32:20
going to get around for VCs and I'm like,
32:22
no, this is really not what you want to do.
32:24
You don't have high costs,
32:26
capitalization costs that you need to pay up front and you don't need huge numbers
32:31
of people.
32:32
Let's try to fund this ourselves or find small angels or even loans
32:38
that you can pay back because your deployments will actually generate some revenue
32:42
and you can fund it.
32:43
So it is important,
32:44
to have the right kind of investors or to match your investors
32:47
with what you're trying to do.
32:48
SPEAKER_3: If you're trying to build a unicorn, by God, go for it and jump in and get VCs and go.
32:53
And that's great because you're aligned with their expectations But that's not true
32:57
for everything.
32:58
SPEAKER_3: I have another guy who's using AI in a super amazingly cool way.
33:03
There are fish farms everywhere right now in Norway,
33:06
in particular with salmon and stuff and Hawaii and stuff.
33:09
SPEAKER_3: They have challenges with the fish growing.
33:12
SPEAKER_3: They don't know how fast they're growing and whether they're healthy.
33:14
SPEAKER_3: They need to know that they're healthy.
33:16
And right now there's no way to do it other than stick cameras down there
33:19
and go look and or go catch some fish and try to figure it out.
33:23
SPEAKER_3: It's very, very manual and not very good.
33:25
So what he did is he used 3D technology from Hollywood,
33:28
because that's where he was originally involved and cameras to monitor the motion of
33:35
the fish and come up with AI models to understand exactly what the fish's weight is,
33:41
and then to be able to detect whether there are any mites or other problems on
33:46
the fish that indicate illness.
33:48
And so by putting these in and automating them,
33:51
it provides an instant report to the people managing the fish stock.
33:55
So they know exactly which fish are doing well
33:58
whether they should be feeding their fish more or less or the same,
34:02
whether they need a treatment or not,
34:03
And then they have these cool laser things that go zap the mites and make
34:07
the fish healthy.
34:08
SPEAKER_3: It's just incredible stuff.
34:10
SPEAKER_3: And he's doing this all, almost by himself.
34:12
It's remarkable how people are coming up with new ways to use technology plus AI
34:17
to solve problems in the natural world.
34:20
SPEAKER_1: I want to switch, Steve, to past and present.
34:24
Are there a list of do's and don'ts that are in the back of your mind
34:29
from your General Magic experience that you apply directly to
34:33
the companies you're working with these days?
34:36
SPEAKER_3: Well yeah, I I'm gonna start with the do's because don'ts are easy.
34:39
The things that I learned from General Magic that I take with me,
34:42
one of them is to really believe in what you're doing and to believe in it for good
34:46
and valid reasons.
34:48
SPEAKER_3: I think it just makes every day and every night that you're working on it a joy.
34:53
SPEAKER_3: We just worked so hard and so intensely and we challenged each other.
34:58
SPEAKER_3: We didn't know what we were doing, but we knew what we wanted to do.
35:01
SPEAKER_3: And so we pushed really hard.
35:03
SPEAKER_3: And sometimes we pushed in the wrong direction.
35:05
SPEAKER_3: And we were working all the time.
35:07
SPEAKER_3: We're working nights, we're working weekends, we were working anytime we were awake.
35:10
SPEAKER_3: Sometimes when I was sleeping, I would come up with ideas and wake up.
35:13
SPEAKER_3: And that was very exciting.
35:16
SPEAKER_3: And it's a moment in time that worked so well.
35:19
SPEAKER_3: And you can't duplicate it, but you can find times that rhyme with it.
35:25
SPEAKER_3: And I find it now with my companies.
35:27
I get so excited when they have a breakthrough and they get things working,
35:32
or they get a customer who really loves what they're doing,
35:35
or they sign a contract and stuff like that.
35:37
And when I see...
35:39
Know sometimes I'll facilitate meetings,
35:41
I'll facilitate strategy sessions and stuff,
35:44
and I'll be,
35:45
gently directing, and then all of a sudden you start to see the energy, right?
35:50
That one person talks to another person and they just get locked in and
35:54
then they start going,
35:55
and then other people join in,
35:57
and it just makes me feel so good because it's just like General Magic.
36:00
So that's what I look for in companies is I look for people who are doing things
36:05
that are that they really believe for the right reasons.
36:08
SPEAKER_3: So that's one thing that I got out of General Magic.
36:11
SPEAKER_3: And the second thing is you can't prejudge people.
36:14
SPEAKER_3: You can try and people do.
36:16
SPEAKER_3: But you need to get in a room with somebody and work with them.
36:20
SPEAKER_3: And then you understand what they're all about.
36:22
You look at them,
36:24
you listen to them,
36:25
you interview them,
36:27
whatever,
36:27
you can come to some understanding,
36:29
but you don't really know a person until you're faced with a real challenge,
36:34
a hard problem, and you both know that you've got to solve it somehow.
36:38
SPEAKER_3: And then you see what that person's made of and you see how they behave.
36:42
So I think that General Magic gave me this perspective where I really expect
36:49
the best of people and I give them room to go do that.
36:54
SPEAKER_3: And if they can't, okay, fine, they can't.
36:56
SPEAKER_3: But I still expect them to.
36:59
SPEAKER_3: And a lot of times they rise to that occasion.
37:03
And so you give people that room and you give them the grace and you give them
37:08
the tools and the resources and give them the support.
37:10
SPEAKER_3: And then after that, then you can judge them.
37:13
SPEAKER_3: But give them the chance.
37:15
I saw people do amazing stuff at General Magic,
37:18
stuff that you can't possibly imagine that they were really capable of doing.
37:22
SPEAKER_3: And so that's another thing that I learned from it.
37:25
And then another thing I learned is that you can take on too much,
37:29
and so it is really important to say no to things.
37:32
SPEAKER_3: And so I tell my companies this all the time.
37:35
SPEAKER_3: You need to figure out what you're saying no to.
37:37
SPEAKER_3: If you say yes to this and this, then you got to say no to some other stuff.
37:41
Because if you aren't able to focus,
37:44
then that tells me that you don't have a clear understanding of what you're trying
37:47
to achieve.
37:48
Because if you're trying to achieve something,
37:50
you you've got to be able to say this fits and this doesn't.
37:54
SPEAKER_3: Where are those boundaries?
37:56
SPEAKER_3: And that's really hard to do, especially with a small company, but you have to do it.
37:59
SPEAKER_3: And you do it.
38:01
SPEAKER_3: Maybe you're wrong.
38:02
SPEAKER_3: That's possible.
38:03
SPEAKER_3: That's why the word pivot exists, right?
38:05
SPEAKER_3: You go, okay, that was wrong.
38:07
SPEAKER_3: Let's switch.
38:09
I think I already knew this coming into General Magic,
38:12
but it really drove it home for me you really do have to have people who are looking
38:17
at the horizon.
38:19
And keeping track of where you are and where you need to be,
38:22
and then effectively communicating that to everyone in the company so
38:27
that everyone else makes decisions consistent with where you are and where you need
38:32
to be.
38:32
SPEAKER_3: When you manage, you don't tell people what to do.
38:35
But what you need to do is give them a clear path to where you need to go and
38:40
where they need to go.
38:41
SPEAKER_3: And let them use their smarts and their resources and stuff to figure out how to get there.
38:46
SPEAKER_3: But make sure that they're on the right path.
38:48
SPEAKER_3: And that's something which, I would argue General Magic did not do a good job of.
38:54
SPEAKER_3: And I try to do that with each of my companies.
38:57
Another way I phrase it is when I'm talking to a CEO or I'm talking to
39:00
a senior executive, I say,
39:03
each person in this company has a unique value,
39:05
a unique role to play in the company.
39:08
Your unique role as an executive is you've got to see the big picture and you have
39:14
to translate it effectively so that the people below you understand it
39:17
and understand how to make decisions consistent with it.
39:20
SPEAKER_3: That's your role.
39:21
SPEAKER_3: And if you're not doing that, you're not doing your job.
39:24
SPEAKER_3: So that's another thing that I think General Magic drove home the point on.
39:29
What do you think around AI and also around the types of companies with
39:34
which you're involved, people aren't thinking about that they oughta be.
39:39
One of the critiques of General Magic is that we didn't think
39:43
about surveillance capitalism
39:45
or how digital communications could actually drive political divisions you could
39:51
hide behind your keyboard and whatnot.
39:53
SPEAKER_1: Are there things that particularly concern you that you think are not getting scrutiny?
39:59
SPEAKER_3: Yeah.
39:59
So let me just say a couple of things about the criticisms about General Magic,
40:03
and then I'll answer your question on AI.
40:05
SPEAKER_3: I think those criticisms, sure, you can make them and they're true.
40:08
SPEAKER_3: They weren't on our radar.
40:10
But I believe that if we had succeeded in getting our products and services out into
40:15
the market,
40:16
I think we had tools and resources that could have been used to address some
40:21
of those concerns.
40:22
SPEAKER_3: Now, there's nothing that can stop people from using technology for evil.
40:27
You can create nuclear fission and you can say you can power a power plant,
40:31
but you can also build a bomb and you can't stop that.
40:34
SPEAKER_3: So some of that is going to happen no matter what.
40:38
And I guess the one other comment I would add is that at General Magic,
40:42
we tried to do so much inventing of so many things that didn't exist yet
40:48
and I just don't think it's possible to see everything in the future.
40:53
It was remarkable that we even saw the personal digital assistant merging with
41:00
the phone,
41:01
merging with a wireless network,
41:04
merging with an internet, and providing universal communications capabilities.
41:10
SPEAKER_3: People today have no idea, but that concept is so far ahead of its time that it's mind-boggling.
41:18
So on AI in particular,
41:23
the biggest problem I have with it is my concerns about confidentiality and privacy.
41:28
Right now China is a really interesting country in the sense that it has
41:35
a multi-thousand year history, a massive timeline of civilization and invention.
41:43
SPEAKER_3: It's a very proud people.
41:46
And they have taken a country that was truly impoverished and backwards and
41:52
in just very short number of years turned it into an economic powerhouse.
41:59
SPEAKER_3: They had horrible pollution problems.
42:02
SPEAKER_3: They've wiped that out in large part.
42:05
SPEAKER_3: They've generated a massive amount of power to power their country and stuff like that.
42:09
SPEAKER_3: It's an amazing success story at that level.
42:13
SPEAKER_3: And they've done it by wiping out personal freedom, privacy.
42:19
SPEAKER_3: And basically creating a powerful dictatorship.
42:23
SPEAKER_3: And I have to tell you it's a really interesting problem.
42:26
SPEAKER_3: You can say it's bad, but you can also say, wow, it's created an amazing economic opportunity.
42:31
SPEAKER_3: I know so many people, right?
42:32
SPEAKER_3: I had employees in China and I saw this firsthand.
42:36
SPEAKER_3: The amazing increase in quality of life.
42:41
SPEAKER_3: In terms of treatment of people and so forth, just it's an amazing breakthrough.
42:46
And I talk to my friends and colleagues who I mentor in India right now,
42:51
and they look at China and they go, hey,
42:53
they're the same size as us,
42:54
same number of people, they've fixed all these problems that we still have.
42:59
SPEAKER_3: Maybe their system isn't so bad.
43:01
So it's before you slam China and say how horrible they are,
43:05
which I'm about to do, you have to say, well, wait a minute, there is value to that.
43:11
But then having said that,
43:12
they've done it by completely destroying personal freedom and liberty and privacy.
43:17
SPEAKER_3: And I am very concerned about that because in a sense it's worse because they were successful.
43:24
SPEAKER_3: They are successful in creating a working country that is top tier.
43:30
I think it's very scary when people can't speak their minds
43:34
and can't do what they want and go where they want to and that sort of thing.
43:37
SPEAKER_3: But that's what's happening there.
43:39
SPEAKER_3: And so I think AI is front and center in that.
43:42
SPEAKER_3: China could not have done what they did without AI, and they're using it very, very effectively.
43:48
SPEAKER_3: And we'll see what happens.
43:51
And so the other problem with AI,
43:54
of course,
43:54
is the same problem that we have with all other technology,
43:57
including, social media or nuclear power.
44:02
It's that we advance technology and then We advance it faster than our culture
44:08
and our philosophy can handle it.
44:10
Things used to move very slowly and people could write philosophical treaties
44:14
on this, that, and the other thing.
44:16
SPEAKER_3: And then we could all discuss it in a community and think about what we think.
44:21
SPEAKER_3: That doesn't happen anymore.
44:22
SPEAKER_3: People don't write, people don't read, and people are, off doing whatever they're doing.
44:26
SPEAKER_3: So I think AI is it's an accelerant.
44:29
SPEAKER_3: It makes things happen much faster.
44:32
SPEAKER_3: And people are really struggling to adapt to it.
44:36
And so what I am concerned about is that it's going to yet again create a dichotomy
44:42
or divergence within the society where there will be people such as me
44:47
who have embraced the technology and are using it in all of its variant ways.
44:52
And that enables me to do a whole bunch of stuff that a lot of people
44:55
who either don't have the availability or accessibility or desire or whatever
45:02
to use AI, they're not going to be able to do that.
45:05
SPEAKER_3: And so it creates different classes of people.
45:08
SPEAKER_3: And I think that's not good.
45:10
SPEAKER_3: Now there are ways for people to use AI to help all of society.
45:16
And I want people to do that,
45:17
but I'm not foolish enough to think that everybody is going to do that.
45:21
SPEAKER_3: So I don't know how that's all going to break out.
45:23
SPEAKER_3: But I'm very concerned about it.
45:25
SPEAKER_3: But at the same time, I know that I have the time and the ability to use it.
45:31
SPEAKER_3: So I need to understand it as much as I possibly can.
45:35
SPEAKER_3: I view it as good.
45:36
Maybe other people don't think what I'm doing is good,
45:38
but I'm gonna try to use it for good as much as I can and then try
45:42
to help other people do that as well.
45:44
SPEAKER_3: And I guess we'll see.
45:46
But I think as with all things in humanity,
45:49
There's gonna be good and there's gonna be bad and we're just gonna have to see
45:53
where it shakes out.
45:55
Well, one of the things I hear you articulating is,
45:57
It's not about a specific action in terms of guarding against these dangers,
46:01
It's an ongoing process, and it's being able to be responsive to what's unfolding.
46:06
And your headlights shine down the road a little bit,
46:09
but you also have a steering wheel and you need You need to use both.
46:14
SPEAKER_3: My view now is very similar to my view at General Magic.
46:17
SPEAKER_3: You can call it naive if you want, but at General Magic, we had a belief.
46:22
SPEAKER_3: And this is why I joined General Magic.
46:25
We had a belief that if we could just get everybody to talk to one another,
46:29
then the world would be a better place.
46:31
Let's create an internet,
46:32
let's create devices that enable us to talk to people in other countries
46:35
and other places.
46:36
Let's translate things and let's make sure that everybody can communicate
46:40
and things would be better.
46:42
SPEAKER_3: It turns out it's not so simple.
46:44
SPEAKER_3: I understand that, but I still believe it.
46:46
SPEAKER_3: I still believe it, right?
46:47
SPEAKER_3: I travel aggressively throughout the world.
46:50
And when I go,
46:51
and I go to,
46:52
Uzbekistan or Eastern Turkey near Iran or wherever I've been recently and I talk
46:57
to actual people are nice, people are good.
47:01
People just want to live their lives,
47:03
they want to raise their families,
47:05
they want to have an opportunity to exist and have fun to some extent,
47:11
but mostly to live.
47:12
SPEAKER_3: And That's true.
47:14
SPEAKER_3: It's true in Africa.
47:15
SPEAKER_3: It's true in Australia.
47:16
SPEAKER_3: It's true in Asia.
47:17
SPEAKER_3: It's true everywhere.
47:18
And so I really believe that pushing technology forward and trying to get people
47:25
to be able to communicate with each other is a good thing.
47:29
SPEAKER_3: It will not fix all the problems.
47:31
SPEAKER_3: It will never make whatever small percentage of people are truly bad, they're bad, fine.
47:37
But For the vast majority of people,
47:39
they're good, they just want to live and they just want to do stuff.
47:42
SPEAKER_3: I just got back from Turkey.
47:43
And the ability to take my iPhone out and put it in that mode
47:48
where you talk one language,
47:50
they talk the other language,
47:51
and it flips it so that my words are translated to their language on their side of
47:56
the phone,
47:56
and their language is translated to my words on my side of the phone is just
48:01
a massive breakthrough.
48:02
SPEAKER_3: Right.
48:03
It enables us to have full conversations with just random people that you meet on
48:08
the street or that are in a restaurant or whatever.
48:12
SPEAKER_3: And this is great.
48:13
This is good because when you re when you stop tribalizing and stop thinking
48:20
about us versus them and we versus they,
48:23
and you start realizing that people are really fundamentally the same in Botswana
48:30
and Turkmenistan and, Israel and Palestine, all this other stuff, right?
48:35
SPEAKER_3: Where I've spent a fair amount of time also working.
48:38
SPEAKER_3: It breaks through a lot of the conflict.
48:40
And what drives the conflict the most are bad leaders,
48:44
who are a relatively small number of people.
48:46
So there aren't a lot of good leaders,
48:48
but When good leaders come out, they make a huge difference.
48:51
It's complicated because you want to feel safe,
48:54
You want to be able to get food, get housing, get education, get healthcare.
48:59
SPEAKER_3: These things matter to people also.
49:01
But fundamentally,
49:03
if we can just make those things available to more people and get people
49:08
to treat other people as humans and the same species as opposed to
49:13
a different tribe, that's a good thing.
49:15
SPEAKER_3: And I still want that to happen.
49:18
And I see examples of technology,
49:21
some of the some of which we started at General Magic,
49:23
making that possible, so I'm optimistic.
49:27
Now,
49:27
I create that reality for myself because I work with all these people who are smart
49:32
and energetic and creating good things that help fix problems and help make
49:37
the world better or help make people's lives better.
49:40
SPEAKER_3: But there are a lot of people out there like that, and I just think we need to support them.
49:45
They're a force for good,
49:46
they're a force for making lives better for a lot of people around the world.
49:51
SPEAKER_3: And those people just need more airtime.
49:53
SPEAKER_3: They need more airtime, they need more resources, they need more support.
49:57
SPEAKER_3: I I'm optimistic because those people exist.
50:01
SPEAKER_3: And they continue to exist no matter how negative and messed up things get.
50:06
SPEAKER_3: So that's my perspective...
50:10
SPEAKER_0: That's a great note to end on.
50:12
SPEAKER_0: Thank you, Steve.
50:13
SPEAKER_0: It was great to talk to you.
50:14
SPEAKER_0: And to learn about what you're doing.
50:15
SPEAKER_0: A lot of really good stuff.
50:19
SPEAKER_2: Thank you for listening to General Magic and the Soul of Tech podcast.
50:22
Tune into our Substack for more interviews and articles
50:26
at spellboundproductions.substack.com.