0:00 So this last session, I know it's the end of the day and everybody's tired. You're
0:03 probably tired of hearing me talk. So I'll try to talk lower because I understand a
0:07 lower voice is a little more pleasing. I use my Samoan voice. Yeah.
0:14 So anyway, is that more pleasing? A lot of vowels, right? Rhythmic flow. Polynesian
0:20 languages are kind of nice. Anyway, so I have some slides, but I actually wanted
0:28 this to be a little bit more of a discussion because this is the problem
0:34 that we're trying to solve as a KERI community. And unlike some of the IAW
0:39 sessions, we get a lot of people that are anti-KERI. Everybody here is pretty much
0:43 KERI friendly, right? So, what is open infrastructure? Have you guys heard that
0:51 phrase before? Open infrastructure? Okay. So, there may be more than one definition.
0:57 This is how I define it. It's based on open standards, open source. But what
1:04 makes it infrastructure is the infrastructure is portable. It allows
1:08 portability by the users. So if you have vendor lock-in, it's not open
1:13 infrastructure. There are lots of infrastructure systems that are based on
1:18 open standards. They're based on open source, but they allow vendor lock-in. So
1:24 they're not open infrastructure. So for example, if you have have a mail server is
1:32 SMTP
1:35 mail. Is that open standard? Are there open source implementations of it? Can you
1:44 easily port your mail from say one
1:49 mail server to another? Theoretically, right? Right? Yeah. Does your mail client
1:59 let you do that? Not usually. Huh?
2:07 Yeah, right. You have to do a lot. It's not easily. Right. Okay. So it's not. So
2:13 despite those two, the portability thing is a problem. Now, why isn't stuff built
2:20 portable? Because the incentive structure is wrong. So to be open infrastructure, it
2:27 should be designed so that the incentive structure fosters cooperation and
2:33 portability. What do I mean by cooperation? Have you guys heard that
2:36 phrase before? It's people are able to make money in a competitive environment
2:41 where they compete on their implementation quality, but there's an incentive for them
2:48 to cooperate so that everybody benefits if there's cooperation. In other words, the
2:55 the network effect is that if we cooperate, we make more money than if we
3:00 don't cooperate. So, it incentivizes cooperation. So, for example, in KERI,
3:06 there's an incentive for people running watcher networks to cooperate. Can
3:13 somebody tell me what that incentive is? Why would one watcher share information
3:17 with another watcher? – Detect duplicity. – Yes, well, they all detect duplicity.
3:24 But what What incentive do I have to get stuff from another watcher and what
3:33 incentive does the other watcher have to share that with me? Ground source and
3:38 reputation. Yes. Right. A watcher is most effective if it has maximized scope. If I
3:47 can see the whole internet, then nobody can eclipse me. Right? Well, how do I see
3:56 the whole internet? If I don't have the resources to physically see everybody's
4:02 witnesses, but somebody has resources to see witnesses that they have an incentive
4:08 to see, and if I have an incentive to see witnesses that they don't have the same
4:12 incentive to see, to spend the resources, and I say, well, I'll give you my, I'll
4:17 give you the witnesses I watch if you give me the witnesses you watch. Do you have a,
4:22 is there an opportunity to cooperate? Absolutely. So if I'm building witnesses
4:26 for the GLEIF ecosystem, I have an incentive to watch the GLEIF ecosystem. I
4:30 don't have an incentive to watch the Utah State ecosystem. But if the Utah State
4:35 ecosystem is watching the state ecosystem, the delta, the marginal cost to share with
4:41 the GLEIF ecosystem becomes small for both. So if marginal costs are lowered and
4:48 sharing is just a marginal cost, then there's a large incentive to share because
4:53 now I've doubled the scope for a marginal cost. Whereas if I had to watch both
4:57 ecosystems, I'd have to spend much more if I had to watch it myself. Right? So
5:02 anytime you do an incentive structure, if the marginal cost is
5:10 lower for cooperation than not to cooperate, then you incentivize
5:19 cooperation. So you build things so that marginal cost is the side of marginal
5:25 savings is on the side of cooperation. So watcher networks are inherently mutually
5:31 incentivized for cooperation because the their value comes from scope and scope if
5:36 I have to watch it myself is expensive. But if somebody else is watching it and
5:40 they share it with me, all they're sending me is a bunch of proofs. They're just
5:44 sending me a bunch of proofs of duplicity. Because they don't have to share me
5:49 everything they watch. They just have to share with me duplicity that they found
5:53 because that's all I care about. So I can say, hey, do you have any proofs of
5:58 duplicity? And they go, yeah, I got five of them. That's all I have to share with
6:01 me. But to find those five proofs, they had to watch a million witnesses. Right?
6:07 And they can say the same thing. How many witnesses are you watching? I'm watching a
6:10 million. Are you watching the same million? No. How many proofs of duplicity
6:13 do you have? 10. Here you go. That's a marginal advantage that now means that
6:20 everybody in their network, every watcher in their network is effectively providing
6:26 coverage over 2 million witnesses. And everybody in my network is getting
6:30 coverage of 2 million networks, but we only have to watch 1 million. So we don't
6:35 have to incur the cost of watching 1 million witnesses, but we get a thing. And
6:39 guess what? That scales because the amount of duplicity is a small fraction of the
6:43 total amount of traffic. So if I have a if every ecosystem shares with every other
6:51 ecosystem, then you get global coverage and the marginal cost is the cost of
6:56 duplicity, not the cost of watching. Does that make sense? Okay, so that's an
7:02 intentional design, right? By design, the watcher network is designed because people
7:07 would say, well, in the blockchain world, well, you have to have incentive for
7:10 infrastructure. So, you have to have tokens, right? You have to be able to to
7:14 mint tokens because otherwise, why would you pay for it? Well, in this
7:17 infrastructure, if you have incentive to watch your local ecosystem, the duplicity
7:24 detection lowers the marginal cost that it's almost free to share what you watch
7:30 when all you have to share is duplicity with other ecosystems. So then it flips
7:35 the incentive structure. You don't have to build expensive token-based incentive
7:39 structures to incentivize people to do it. So the problem is that nobody makes money
7:48 in infrastructure that's open and and portable until you reach scale. Nobody
7:54 makes money in it, right? So you have what's called a bootstrap or cold start
7:58 problem.
8:01 So we've kind of talked about this before, but the everything in KERI is designed so
8:10 that there is open decentralized centralized, not
8:20 shared infrastructure.
8:29 So you have shared data,
8:35 but no shared governance.
8:42 Okay, that's the that's a critical property. What does that mean to have
8:46 shared data and not shared governance? Well, if I send data to you that you can
8:50 independently verify without having to trust me, then you don't care. You can
8:55 verify it. You don't care where it came from. All data is created equal as far as
8:59 you're concerned, right? Shared governance means that I can't trust it unless I trust
9:05 the governance. So in a blockchain, everybody has to trust the governance.
9:10 They have to trust the… Usually the governance is a distributed consensus
9:13 algorithm and the token structure that incentivizes people to run nodes in the
9:19 distributed consensus algorithm. And so the tokenomics of blockchain are critical
9:24 to that shared governance working. KERI says, I can't afford that. That's too
9:28 expensive. Shared governance is extremely expensive. So what I need to do is make it
9:34 so that it's truly totally zero trust. So it so I don't need shared governance
9:39 because everybody can independently verify it and I need to make that independent
9:42 verification cheap enough that they afford to do it. but it's not zero. That's the
9:46 problem. It's still not zero. We've reduced the cost floor as low as possible.
9:51 So the cost floor is as low as we can make it because we got rid of shared
9:55 governance. We made everything independently verified, but it's not zero.
9:59 So we still have the problem. We have a cold start bootstrap problem. How do we
10:03 get enough infrastructure to enable scale and the people providing the
10:09 infrastructure to be able to afford to provide it to until it reaches scale so
10:14 that that they can make money because they won't make money until it's at scale. If
10:18 I'm running a witness service, I'm not going to make money until it's at scale.
10:21 Because I'm not going to charge people, you know, $100 a month for a witness,
10:26 right? Nobody could afford that. They want to pay a dollar a month or maybe 10 cents
10:32 a month, right? So, how do I afford it? That is the hard
10:41 problem. So what we're trying to do in Utah in SEDI is make the case that if the
10:52 state of Utah primes the pump, everybody else that wants to be in that ecosystem is
11:01 incentivized. So a public identity
11:12 Utility. That's the term that we came up with. So what is a utility in a regulated
11:21 environment? What's a regulated utility? – Electricity. – Electricity. Okay. Power. –
11:29 Garbage. – Telecom. Right? Garbage. Sewers. Right? And the reason why do we
11:37 have utilities? Now, if you're from the state of Texas, you probably are going, we
11:43 know why you have utilities. Because the state of Texas has a mishmash of a power
11:49 grid because they didn't create regulated power companies. But if you're in most
11:55 states, there's only one power company. The power company is a regulated utility.
12:00 It's a public-private partnership. They get to make money, but they can't make
12:07 money without about getting approval for any rate increases, any changes in
12:14 infrastructure. So they're guaranteed to make a fractional profit, which if you're
12:20 big enough, a teeny bit compounded over years amounts to a lot of value, but they
12:27 get a monopoly, right? So they get a monopoly, so they don't have to compete
12:32 with anybody, but in return, you get to regulate them. So they're not the most,
12:36 they're not, they don't, you're not guaranteed the most efficient, but you do
12:40 guarantee that you don't have non -portability because everybody's power is
12:46 portable, right? It doesn't matter which house you go to, power just works, right?
12:51 So a public identity utility says you're going to have an identity system that is
12:58 portable and interoperable everywhere, not because you've forced people to
13:04 interoperate, but because you basically said, we're going to have one. We're going
13:08 to have a utility. It's going to be an identity system and the state is going to
13:14 regulate it and
13:19 they're going to standardize it and they're going to ensure that there's
13:30 enough money to build out the infrastructure. Sure.
13:37 Okay. So that bootstraps it. So what are the bootstraps? One is IAL 3. That's
13:48 identity assurance.
13:52 Okay. Is the state of Utah or all states in the identity assurance business
13:57 already? They already are. They already do it. They don't do it to IAL level 3
14:02 because they don't have to to. But if I'm going to have a persistent identity that
14:07 lasts a lifetime, I want to make sure that when I start it, I start it with a really
14:14 high level of identity assurance because guess what? With KERI, I can amortize that
14:20 one identity assurance session across the lifetime of that one citizen. No other
14:26 identity system allows you to do that. Every other identity system, every time
14:29 you rotate keys, you either reincur this cost or you're insecure. And guess what
14:35 they've chosen to do? Not re-incur this cost. Which means the first time you
14:40 rotate your keys, you've now become insecure. That's that describes the
14:43 certificate authority system. CAAs are supposed to repeat the level identity
14:50 assurance every time they reissue a X-49 certificate. That's what the spec says. Do
14:56 they do that? How many people have have issued new certificates on CAAs? How many
15:05 of you actually went through any kind of identity assurance for any reissuance of a
15:09 certificate? That's not what the spec says. Certificate authorities are supposed
15:14 to redo, re-verify everything about every piece of information in that certificate
15:19 when they reissue it. What's that called? Let's encrypt? Yeah. Okay. Right. Okay. So
15:28 that's the problem is is that if you can't amortize this cost, you're incentivizing
15:34 insecurity. So KERI allows you to amortize that cost and the state in SEDI is going
15:42 to pay that cost.
15:48 Okay, but that isn't the infrastructure. That's just bootstrapping the identities
15:53 that need the infrastructure. I'm not sure how they're going to pay for the
15:56 infrastructure. There's been discussions about them paying for infrastructure for
16:01 disadvantaged populations where the state will actually subsidize infrastructure
16:07 providers to provide witnesses and watchers by issuing coupons to
16:13 disadvantaged population to be able to get accounts on those services. So that's one
16:18 way to bootstrap it. Okay. Another way to bootstrap it would be for providers of
16:27 witnesses and watcher services to have have a market opportunity that allows them
16:36 to make money on something else that is enabled by them giving away witness and
16:44 watcher services. So if I'm a bank and
16:53 the state and the federal government, this is a big if, but I think they just passed
17:01 a law allowing it. I think that just happened. That allows one bank's KYC
17:14 to be used by another bank. What did I just do? I lowered the cost for every bank
17:23 because every time a customer changes banks right now that bank has to incur the
17:30 full cost of the KYC every time. So portability is expensive. Well that and
17:36 the banks have to do it. And the thing is they can't really use it to stop the
17:39 customers from moving because they have to do it one way or the other and the
17:43 customer can just say, you know, I want to transfer them. I want to move to a new
17:46 bank. And the bank says, okay, we have to move your money, but we're only going to
17:51 move it to another bank. And that other bank won't ask us to move it because we
17:55 can trust that the other bank is following the regulations or there wouldn't be a
17:58 bank. So there's already enough trust that you can move your money between banks, but
18:02 every bank every time you move your account is having to incur the cost of
18:06 KYC. Well, if you have an IAL level three and a regulation that allows one IAL level
18:13 three to move between banks, then all you have to do as a customer is issue an
18:23 authorization to the new bank to use your existing identity in the IAL level three
18:30 that the state paid for to do that. So now the banking industry has an incentive to
18:36 issue to create free witnesses and watchers because they get KYC savings.
18:44 Because your portable KERI SEDI endorsed identity now becomes a savings to them. So
18:53 that's an example of how if you do this right, if you think it through, you can
19:00 start start to build incentives, right? And so if you have to have witnesses and
19:08 watchers talk to the state and the state says, well, we don't want to pay the full
19:15 freight for witnesses and watchers because we have lots of citizens who make enough
19:18 money and so we want to impose on them a hidden tax, right? States do this all the
19:26 time. So you're going to have to pay for witnesses and watchers, right? But we'll
19:30 pay for disadvantaged populations. But everybody in the state also has a bank
19:35 account and the banks say, well, wait a minute. If we just spin up witnesses and
19:40 watchers for our customers, then we can get free KYC because the state paid for
19:47 our KYC and we didn't have to pay for it. And if you look at the cost to the bank of
19:51 the KYC, you are in the banking industry, how much does it cost to onboard one
19:54 person for KYC? What does it cost the bank? I don't know what the exact figure
19:59 is, but just multiply it for many times within the bank because you have to KYC
20:04 yourself whether you're a person or an organization for every service that you
20:09 apply to. – Every service you apply to. So let's say, could it be $100 a year? Okay,
20:16 well, $100 a year pays for some really awesome web scale infrastructure. I mean,
20:23 how many witnesses can we run at $100 a year? – It's a year. I couldn't even
20:29 guess. but it's, you can run a lot. – Right. So, I mean, we have four droplets
20:34 that cost us about 100 bucks a month for four droplets on DigitalOcean.
20:41 Each droplet has, I think, is a four or eight cores. I can't remember. – Four
20:46 cores each. – Four cores. A ton load of space, quite a bit of stuff. So they could
20:53 easily run 10,000 witnesses and 10,000 watchers and not not burp. And that's a
21:01 hundred bucks a month, right? So if a bank wants to run a witness and watcher service
21:07 for a hundred percent of their customers, it's $5,000 a year max. They're spending a
21:16 hundred times that on KYC. So if they do the numbers, they go, you know what? We
21:21 should just spin up free witnesses and watchers to incentivize people to KYC us
21:26 using their SEDI identity instead of KYC in the old-fashioned way where we have to
21:32 do we have to carry the freight, right? Every industry that does identity has a
21:38 similar cost trade-off where this identity assurance is the savings that the state
21:48 provides by giving them portable identity assurance. Right now, identity assurance
21:51 is not portable. Without KERI, it's not portable. Everybody has to redo it. But
21:55 with KERI, it becomes portable. That they should be able to justify the cost of
22:00 spinning up free witnesses and watchers just to so that they can gain that savings
22:05 for their customers. That's the idea of this. That's the motivating factor for the
22:12 state funding that is to be able to and the thing and for this to work, a bank has
22:19 to be able to say, well, I've got an IT person at the bank. Show me how I would
22:23 spin up free witnesses and watcher services. because I don't want to pay
22:28 somebody to do it. I just want to download this open source, open standard software
22:33 and just run it on my own servers, right? That's what the KERI Foundation's
22:37 attempting to do is build enough open source code to set up so that people like
22:44 a bank can go and do it. What other industries can you think of that benefit
22:48 from onboarding?
22:52 – Insurance. – Insurance. Absolutely. Absolutely.
23:00 You know, one of the most fraud-filled industries in the world is right now?
23:09 Insurance claims.
23:14 And why is that? Because I can deep fake an insurance claim, especially in the
23:22 healthcare industry. It's easy. I know what's on the form because the forms are
23:28 public. I know your address. I know the address of all of your providers that
23:33 submit insurance claims. And you don't have a good way of KYC-ing the submitters
23:39 of the insurance claims.
23:44 So every day you'll read about, oh yeah, we lost a million dollars because this
23:50 fraudulent pretended to be a clinic someplace and submitted all these
23:56 insurance claims for all of this healthcare work with and they and they had
24:00 all the the right numbers the claim numbers and stuff like that because they
24:03 stole them from from me the the insurance payer because my IT was so bad that they
24:08 could steal all of that the the that information so they could submit
24:12 fraudulent insurance claims. So insurance is a good one. Another one? Anybody?
24:19 – Any messaging? – Huh? – All kinds of messaging. Yeah, but money. Messagings
24:26 tend to be free. We need something where somebody is making money and they're
24:32 spending a lot of money to combat. Because the only thing the state is providing is
24:38 this. So we have to leverage this free thing the state's doing to incentivize
24:45 people to say, well, to get this, I'm willing to spin up with free witnesses and
24:50 watchers. Yeah. All of employment. All of employment. Yes. Every HR department of
24:57 every company has to KYC their employees to onboard them. And they have to verify
25:04 their transcripts. They have to verify their employment history. They have to
25:10 verify their claims history for prior…
25:17 what is it? – Convictions. – Yeah, prior convictions. Yes. That too. Huh?
25:27 Background checks. Yes, they have to do background checks. They don't want to hire
25:30 somebody who's a criminal, right? Okay. – Government service. – I was going to say
25:36 government service. – There you go. – Like EBT, SNAP, web service. – Yes.
25:47 Well, certainly every Well, okay. So the state, since the state provides those, the
25:54 state has an incentive to use its own IAL level three. We're talking about people
25:59 that aren't the state because the state's going to do that anyway, right?
26:05 Who else? A lot of times now what I'm seeing is that they are sharing that
26:11 information. So like in the state of California. Oh, yeah. Or in my city
26:16 anyway. If you've got, I'm sorry. So like in my city, if you've got PG&E care,
26:24 they'll share that with EBT or EBT will share that with such. So it's not
26:28 necessarily just government. It's because once they've gotten it from one, now you
26:34 can get all these other services more easily. And that's one of the reasons why
26:38 healthcare is one of the adoption vectors for KERI because it's probably the worst
26:43 for that. It's the single biggest sector in the economy and it has the worst fraud.
26:48 of any of them because people have to have healthcare, right? And they have to
26:55 deliver it. They're required to deliver it. And they have to deliver it even if
26:58 they can't really verify the person because they can't turn somebody away.
27:02 Somebody comes in and says, I'm dying. Well, we can't verify all of your stuff.
27:06 They're still going to have to deliver the care, which means they either get paid or
27:10 they don't. And sometimes they just, you know, the payers have to deal with the
27:15 fact that they don't have really good provenance. So all consumer services that
27:20 have to verify something about somebody who buys their services? Yes. Like buying
27:26 cigarettes, buying alcohol, buying… Yeah, every regulated denied
27:36 service, right? So alcohol, pornography, and you know, in the state of Utah, social
27:45 media for 13 plus or under 13 is denied and 13 plus is allowed, right? So the
27:55 point is, is that if you think about it, And you think about the core thing is
28:00 you're getting this thing for free that's never been free before. And I wish Phil
28:07 was here because he wrote two, he wrote a white paper where he said, he said, and I
28:13 didn't prompt him to write it. He said, the thing that's always killed digital
28:16 identity is that the identity assurance was too expensive. Nobody wanted to pay
28:22 for the identity assurance. As soon as somebody like the state pays for identity
28:27 assurance, for individuals, then it unlocks the thing that stops identity from
28:34 digital identity from being used in general, which is now you have the most
28:39 expensive thing that's a gating function paid for. Now with KERI, since you can
28:46 amortize that gating function across the lifetime of an identifier, which means if
28:54 it costs you a hundred bucks to do KYC, that hundred bucks gets amortized across a
28:59 hundred years. Whereas every other identity system can't amortize it. So it's
29:04 even worse for them. Right? If I'm doing DNSCA, every time I churn my certificate,
29:10 I have to re-incurred that KYC. But KERI not because perpetual identity. That's the
29:16 perpetual identity means I get to. So here's the equation. Perpetual identity
29:21 equals IAL3
29:26 forever.
29:29 Okay? You do it once, the cost is amortized across the full lifetime of the
29:36 perpetual identity. So it's the minimum cost
29:43 to bootstrap. It's not zero, but if this is valuable enough, then people should
29:53 bring free witnesses and watchers to enable them, their customers that they're,
30:03 that this is a cost center for, to benefit, for them to benefit from that by
30:09 enabling their customers to use a perpetual identity system. Yes. I think
30:13 there's a lot of synergy too. Like even industries that don't necessarily rely on
30:17 identity assurance right now, they, if it's free, they could, and it would just
30:22 augment in a very synergistic way. like I know in the insurance industry, there's,
30:27 you know, you have to, the insurance companies have approved, you know, people
30:31 who fix things for, for, you know, whatever was damaged. And so that could be
30:37 used by adjusters, right? Yeah, adjusters or even like thing like the auto mechanic
30:41 shops themselves. Oh, yeah. Or the shops that do like home repair and, and things
30:46 like that. They could even use that. Yeah. So there's a bigger ecosystem. So the
30:49 point I'm trying to make is, is that if you're thinking of starting a company,
30:55 with KERI, you're not going to make money spinning up witnesses and watchers. I'm
31:02 sorry. It's just not that only the only the people that can operate at huge scale
31:06 are ever going to make money doing that. But if you can benefit from IAL3 plus
31:13 forever by giving away witnesses and watchers, you have a business model
31:18 because now the infrastructure that stops KERI from working becomes free. KERI
31:26 infrastructure becomes free because the state of Utah is paying for this. That's
31:31 the opportunity. Yeah. – So, what concrete ways have has the state of Utah indicated
31:39 they might be willing to pay for it? Is it just the… I know you mentioned… – No, no.
31:43 They've said they are. They have to. – You can't reach citizen. – Yeah. You can't get
31:47 a driver's license without… If you want a SEDI, you're going to get an IAL level
31:53 level three identity assurance process. Wow. Yeah. Did they say they're going to
31:59 like integrate it into the DMV or something? Yeah. Wow. Or they might hire
32:02 ID.me to do it for them. I see. Or anybody that shows up and says we're good enough
32:07 at doing identity assurance and we want to do it for the state and the state will
32:12 issue them a contract and give them a delegated identifier to be an agent for
32:17 the state, an issuing agent.
32:22 So, So, so the point is, is that that's the place where you can make money in the
32:27 KERI ACDC ecosystem, not infrastructure. You want the infrastructure, you want to
32:33 give it away. You want to essentially, you want to give away KERI infrastructure and
32:38 then make money on the fact that the that you get IAL3 forever and the state paid
32:44 for it. Yes. – So, I think what's making the state want to do this based on what
32:49 you've said today, correct me if I'm wrong, is because of legislation that they
32:53 put in place that had no mechanism for doing the social media stuff for the 13
32:58 and above. – So now they backed themselves into a corner for that. – Yes, they backed
33:02 themselves into a corner. – So this is my, I'm going to look at the magic ball here.
33:07 The consortium that's coming up of multiple states that are going to support
33:12 this. – They have similar problems. – Do you think there's a possibility that there
33:17 would be kind of a cross-state line legislation? – Yes. – Yes. They want to
33:22 have – They have reciprocity. They want to have reciprocity between states. So that
33:28 if one state does IL level 3, that you can move to a different state and not have to
33:34 repeat it. Does that make sense? It's not just forever in your state, it's forever
33:42 in every state. – So, what is the identity proofing instruments that you have to
33:48 produce? – Well, IL level 3 Yeah. Says you have to use so many first class documents.
33:58 That means passport. You know, and then you get so many second class documents and
34:06 then you have to have a biometric and then you have to have a secure way within
34:09 session to bind that proofing session to an identifier. Okay. So, this is a future
34:16 question. Okay. Okay. That's how we would have to back end into something like this.
34:20 But let's go to the future where, you know, the baby that was just born last
34:26 week and they start to get their identity. At what point do they get an identity at a
34:32 state level? Do they have to be a certain age or… – No, no. They could get it at
34:37 birth. And the reason they can get it at birth is that in general, they trust the…
34:49 in general, states trust the chain of custody from the womb to the form that you
34:56 fill out that says this is a person. – They trust the hospital, not you. – Yeah,
35:01 well, I said the chain of custody. – Yeah. – Which is you're in the hospital – Yeah,
35:06 but you said the form you fill out. – To the form you fill out. – Yeah, but you're
35:10 filling out the form of stuff. – Yeah, but the hospital is the one that certifies the
35:14 form. – Yeah, but mom doesn't fill out. – So, so – Yeah, well, okay. The hospital
35:18 fills it out for the mom. Well, actually, birth certificates are done differently in
35:23 every state. So, that's what I'm getting to. So, there will have to be a
35:28 normalization so that states can trust birth certificates from other states.
35:31 That's one of the worst provenanced
35:36 primary identity documents are birth certificates. And George McEwen,
35:46 he likes to give the example of he was in a state And the way they did birth
35:51 certificates was a paper entry in a log. That was a birth certificate from some
35:57 little town where somebody signed it. And even today, it's so, I mean, you look at,
36:03 you know, in California, you want your birth certificate altered. It's not even a
36:08 court process or anything. You just ask for it. Yeah. So, that is, so that will
36:13 require some, so the thing is, it won't happen overnight. Those sorts of things
36:20 will take time to iron out, but yes, that's where it's going to end up. Yeah.
36:26 Yes. There you go. The rumor is that there may be as many as 15 states in the
36:35 consortium. Yes. I've heard it's double digit. That is true. It's about a third of
36:40 the country. That's too much to ignore by pretty much any industry in the identity
36:46 space. So, So if true, which we won't know in until two weeks at the SEDI summit, it
36:56 will be like a bomb went off in the identity industry. It will be a really big
37:01 deal.
37:04 And the fear, I'll tell you the fear, is that they will say KERI and ACDC isn't
37:11 mature enough. And that's a real fear. And what I'm trying to do is I'm trying to
37:16 tell the KERI community, there's a reason to contribute to open infrastructure
37:23 because open infrastructure unlocks all of the business opportunities that aren't
37:31 going to be unlocked any other way because you don't have anybody else in any other
37:38 vertical paying for the IAO level three. But SEDI, you have that forcing function.
37:45 Go ahead. Oh, you have a question? Yeah, I was going to say something. Okay. I think
37:51 we have another mic over here. Yeah, I wanted to give some experience that we've
37:55 been having with the VLEI and in seeing, you know, this situation here that SEDI is
38:05 putting forward or the requirements that SEDI is putting forward, it's only going
38:10 to work because there's there is regulation and legislation behind it and
38:18 the states are behind it because the actual experience that we're having with
38:25 the VLEI is that I'm being told time and time again that even though I say that
38:32 these are long-lived credentials and we're trying to be able to put the work in at
38:39 the beginning in order to be able to make them secure enough to be long-lived
38:44 credentials. I always get thrown back that it's too complicated. It takes too much
38:51 time. It's too costly. You can't take somebody who's a board chair or the CEO of
38:58 a multi-million dollar company sitting down even for two minutes to be able to do
39:03 this. And I'm there like, isn't the threat of compromise because they're important
39:08 people incentive enough for them to want to be able to have this kind of credential
39:14 and still the same arguments keep on getting thrown back because people are
39:20 only familiar with the processes that are not even not being carried out by CAs now.
39:28 So they're comparing already flawed processes to something that's – Yeah,
39:33 they're saying, well, this flawed process is low friction. – Yeah. – Your flawed
39:37 process is high friction. The problem is this is not a flawed process that is, but
39:44 they don't see the advantages in the and and and so it's a challenge. It's a it's a
39:49 hard thing. User experience is a difficult thing to solve. – I would never have a CEO
39:53 be in charge of credentials for a company. – Yeah. – They're unreachable people and
39:58 like I'll have ADHD in the business, right? – I can't hear you with that in my
40:02 online.
40:06 – CEOs are inappropriate people to be in charge of the company's basic
40:10 infrastructure, right? They're not responsible people. They don't usually
40:14 like they you need someone like very responsible like a CFO, a COO like someone
40:20 who is a execution focused someone lowered down even. We're talking about the CEO
40:25 receiving one of these credentials and not taking the time in order to be able to get
40:30 one. So they're the receiver. They're not managing anything about the process. Yeah,
40:35 they shouldn't be involved at all. I don't think. Well, they need to be involved in
40:38 the process to receive their digital identity credential. – But why do they
40:43 need a… – Yeah, so if they go to a board meeting and they want to vote as CEO and
40:48 they want to prove their identity. – They sign regulatory reports. – Yeah, they
40:57 impose liability upon themselves every time they sign their K10 or whatever it is
41:02 for their annual report in the US. – So they don't understand that they the
41:08 liability that they subjected themselves to when they So, yeah, but it's a valid
41:12 point. They don't think that they have to care about that. You had a comment? Oh,
41:19 okay. Sorry. So anyway, I don't have a… When did we end? In 15 minutes? Huh? –
41:25 Yeah. – Two things. One about the complexity of KERI. We've been hearing
41:32 that criticism for years. And I think that that now has to melt away because people
41:39 use a AI to do their coding and handle complex things. And my experience with the
41:47 frontier models is they are big fans of KERI. When you ask hard problems about key
41:52 management and if they don't have to if the model is not completely familiar and
41:57 you give it material to make it familiar, it immediately becomes a fan and all of a
42:01 sudden has answers for a lot of hard questions it didn't have. And if it's
42:05 already familiar, it recommends it. I mean, Zach was here earlier today and the
42:09 reason he's part of the community is because Claude recommended KERI to him.
42:13 And it's the reason why he even knows about KERI is because Claude has that. So
42:18 I think that complexity argument is going to start to go away. Secondly, the cold
42:24 start problem that you referred to earlier. Yes. That's something that's part
42:28 and parcel of any two-sided market, right? Yep. You have supply and demand and you
42:33 try to get the supply going and the demand going at the same time and it's hard to
42:37 juice both parts of the market. With Utah doing what they're doing and potentially
42:42 15 states, if it's could be 10 states, could be five states, might be 20, who
42:47 knows what the real number is, but a good number of states joining them moving
42:50 forward. What they're doing is they're priming the market with demand. Yeah. And
42:55 what we've been waiting for for years and it's coming to IIW and talking about these
43:01 different identity technologies, we've been playing on the supply side. We've
43:05 been playing on, oh, there's this cool thing that we want everybody to want. We
43:09 want everybody to have self-sovereign identity. We want them to have these
43:12 technologies.
43:15 But we couldn't make them want it because we are all staring at the cold start
43:19 problem of the two-sided market. So there is a kind of a once in a lifetime
43:24 opportunity where you have a bunch of states that are banding together
43:28 potentially led by our home state of Utah, your and my home state of Utah, who are
43:33 priming the pump and putting their thumb on the scales on the side of of demand,
43:40 which means money. That means revenue. That means business opportunity for those
43:45 that can come in and provide this supply. So it's going to happen. It's only a
43:50 matter of which companies can come in and provide that supply and satisfy the
43:54 demand, but the demand is absolutely coming. Yeah, I firmly believe that it's
44:00 going to happen, but there is a risk that it doesn't. And if it doesn't happen, then
44:06 it may never happen because continuous identity will be what the market provides.
44:13 And everybody will say, well, we have no choice. They will just say, there isn't
44:20 another option. You know, I think it needs to be said why the state of Utah is moving
44:26 forward because it was speculated a little bit earlier what the motivation is. And
44:30 people need to understand what Utah's motivation is for doing this. It's not
44:35 because they want everybody to have digital identity. It's because of privacy.
44:42 It's their chief privacy officer has gotten everybody motivated and frankly
44:48 scared about the identity options that are that exist. The alternatives are bad. And
44:54 if they don't do something positive, then everyone's going to end up with something
44:57 negative from a privacy and liberty standpoint. So that's the argument that's
45:01 winning. And that's why it's unanimous on both sides of the aisle, left and right. –
45:06 Yeah, because everybody cares about liberty. Completely unanimous. – It's less
45:09 about privacy, more about liberty. – Yeah, it's more about liberty. They've gotten it
45:14 that if you're against this, you're like, hate your mom and hate apple pie and hate
45:18 baseball. And they've really gotten it politically toxic to oppose it from a
45:24 liberty standpoint. And so when you say that it might not happen and that the
45:29 continuous identity of AI, that surveillance identity, is what the market
45:34 will offer, I can almost guarantee you, because you and I have been meeting with
45:38 state of Utah for how many years now? Three. Almost four years. Almost four,
45:42 yeah. Yeah, coming in March, it'll be the fourth anniversary or April of our fateful
45:46 first meeting with them. And Utah, if they have to go it alone and it'd be one state.
45:55 They would do it. Utah would go it alone. Utah would go it alone and be the only
45:58 ones with their own identity technology because they're going to do what's best
46:02 for the citizens of Utah for privacy and liberty even if no one else follows them.
46:07 And others will. I think they'll get other states. But that is the principle. It's
46:11 freedom, it's liberty, and it's privacy. That's what's motivating them. – So, I
46:17 have a question about the consortium. You know, this started with legislation and
46:21 that's what's driving it, not the technology. – That's right. – And then the
46:24 legislation had to find technology that matched what they wrote. – Well, it's both
46:30 ways, right? So, the original social bill, right, was written without any technology.
46:37 Every bill since then has been written with the influence of people from the KERI
46:42 community guiding it, right? So, they know the end point, but they can't jump there.
46:49 They have to say, well, no, we have to lay the groundwork. We know that we want to
46:53 get to there. We know the technology will allow us to get where we're going. When
46:57 they did the social engineering one, they didn't know that. But the next bill, you
47:01 know, that they wrote in 2024 and then in 2025 and then 20 and then 2026, all of
47:06 those were based on they now had a roadmap at that point. – Yeah. So, but this state
47:11 has this legislation which has been developed over, you know, four years,
47:17 right? – Yep. – So, the consortium, is it more going to focus on getting similar
47:23 legislation into other states? – Yeah, they're using Utah as a model legislation.
47:30 – So that, you know, they can go to their states and however their legislative
47:35 process process works. Cause so that to me it's several more years that these states
47:41 will actually have the legislation. – Yeah, they will roll out slowly over but
47:44 they'll go faster than Utah. – Because they have a blueprint to follow. – They
47:47 have a blueprint. So they can jump to the head of the line. – I also think that
47:52 there's something else that another benefit that comes along with this. You
47:59 mentioned earlier today that the paper will not go away. There will be an opt-out
48:05 clause. – Yes. option for citizens. But you have to think that most citizens and
48:12 businesses and the government itself will try to be able to digitalize as many
48:18 services as they possibly can. Security will be up, cost will be down. – But in
48:24 the state, in the SEDI bill in the state, every citizen has a right to opt out. –
48:29 But I'm banking on the fact that most citizens will opt in. – Oh, yeah. – So,
48:35 therefore… – And what The cost of delivering the service. – What will happen
48:38 is that… – It's not opt out. It's opt in. – It's opt in. Yeah, you have to opt in.
48:43 It's a positive opt in. Sorry, I said it wrong, but you're right. – Okay. – They
48:47 have to opt in. – They will choose to get a digital identity. – Yes. – Let's say it
48:51 that way. – And what will happen is demographically as people die off, the opt
48:56 -inners will go up. And at some point in like one generation, you'll have a teeny
49:03 fraction of the population that won't opt in. And so then as a byproduct, a natural
49:08 byproduct as more opt-in, then the cost of services will go down. The ability to
49:15 deliver services faster will go down. And there will be a net positive effect of the
49:25 state being able just to be more efficient. And so that also to the privacy
49:30 and liberty, that should also be pointed out to the other states. I don't know if
49:35 people know this, but for the last 10 years, the state of Utah has been ranked
49:39 as the number one state as a business friendly, well-run state. So they get that
49:46 running the state efficiently matters. And so it's not lost on them that having a
49:52 digital identity system that allows them to run it even more efficiently, but
49:56 protects the liberty of its citizens is part of maintaining their lead over other
50:03 states in terms of how the state is run. And I'm bragging on my home state, but you
50:08 just go look, which states are the best run states? The state of Utah is always
50:12 one or two every year. Yeah, go ahead. I think the adoption is going to be slow and
50:19 then fast. And it's going to be a tipping point that I think will be like tap to
50:23 pay. I remember I was a slow adopter of tap to pay. It took me a couple of years
50:29 before I was willing to do that. I can't remember why if I had any good reason. But
50:34 as soon as I switched, oh my gosh, I would never go back. And so, every place I go, I
50:42 can pull out my credit card, but I moan and groan if I have to go through that
50:46 amazingly difficult effort of pulling out a credit card. And I'm exaggerating, of
50:50 course. But now my wife, she saw me doing it and she was a slow adopter. And now
50:55 she's got her Apple watch and now she just taps her watch and just taps her wrist.
51:00 And now it's like whenever we see someone else pulling out a card when we're doing
51:04 it like the right way, the cool way, the fast way. And now everyone around us is
51:09 tapped to pay and I rarely see someone pull out a card. And it seemed like it was
51:13 slow and then all of a sudden, but there's still always the right to still use the
51:18 card, but the tapped to pay becomes the obvious thing to do. I think that's going
51:22 to happen and I think it'll be a five to 10 year. Yeah, it'll probably be five to
51:25 10 years because we have to work out the user experience and it'll take a while
51:30 just to figure that out. Yeah. Yeah. Pass in the mic. So I was asking you before
51:39 regarding running this open infrastructure at the edge and the problem I was thinking
51:47 to solve was this code stack. Do we have a plan B? Does this sound like a plan B?
51:56 Like growing this infrastructure more organically, more sovereignly? Well, like
52:03 for example, health KERI doesn't need SEDI for it to be successful in the healthcare
52:09 sector. Organizational identity will pay for KERI-like infrastructure in those
52:17 things. It's just that it's a two-sided network problem. And so a startup has a
52:22 lot of stuff they have to do to bootstrap that. It takes time. Right? So that's
52:27 always been the problem. Let's say if we start bundling witnesses on watchers in
52:33 our wallet. – Yeah, absolutely. – And suddenly companies or organizations can
52:38 tap these of services. – Yeah, if forward -looking companies do something like say,
52:46 you know what, we're going to raise, if you have the ability to say raise an extra
52:50 10 million, right? And you say, what we're going to do is we're going to bootstrap by
52:56 scaling, by hyper scaling, right? So how you hyper scale is you give everything
53:00 away, right? So it doesn't cost anything for anybody. You just give everything away
53:05 and you know that you have a runway that says we're going to spend 10 million or in
53:10 some cases 100 million. If you're an AI company, it's 100 billion that you're
53:14 going to give away to bootstrap the adoption. And then once you reach the
53:20 tipping point, then you can start selling tokens, right? ChatGPT didn't cost
53:23 anything for how many years, right? A couple of years. And then all of a sudden
53:27 it's like, oh, well, if you want really good ChatGPT, you're going to buy tokens,
53:31 right? But they spent billions in those first two years. Right? So, if you have
53:37 that kind of resource that you say, you know what? Digital identity that looks
53:42 like this, that it's secure, that solves fraud, that does all those things. We have
53:47 a business model that we'll be able to monetize. You're not going to monetize the
53:50 infrastructure. You're going to monetize all of the services you build on top of
53:54 that infrastructure. Then you would just give it away. You would just say, you know
53:59 what? We're going to go do what the state of Utah is doing. doing. We're going to
54:02 find 10 million people. We're going to give them wallets and we're going to give
54:05 them witnesses and we're going to give them watchers and we're going to then, you
54:10 know, let them play with it. And they'll go, what am I doing with this? And they
54:14 go, well, here's a place you can use it. And it'll be like the tap to pay. Well,
54:18 yeah, maybe I don't want to go there. But at some point, people will find the
54:22 utility and then all of a sudden it'll be like, why weren't we investing in your
54:26 company three years ago when you were giving away millions of dollars worth of
54:29 infrastructure and it looked like bad business model. But you got to have that
54:34 kind of money and I don't have that kind of money. So, yeah. And I don't know how
54:38 to convince people to have that kind of money. But you're welcome. If you can do
54:44 that, I say go for it. – I was thinking of our phones as infrastructure. – Yeah.
54:52 Well, the way I imagine this is presenting credentials from my phone, not from my
54:58 desktop app most of the time. So you will be running this mobile wallet anyways, if
55:04 we could expose this. So the biggest problem is exposing the services running
55:09 in your phone to the outside internet. But yeah, might be doable. Well, every phone
55:17 does have an IP6 static IP address. So you technically could run… …snap networks
55:22 using IPFS or … So technically you could do it. That's great. – Yeah, I think the
55:32 desktop would be a good thing for the CEO from our CEO's conversation. – Yeah. – If
55:36 they had a nice desktop app. – So I'm just offering. Anyway, so before everybody
55:43 leaves, should we close the session? Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire Dire
55:50 Dire Dire Dire Dire Dire Dire Thank you.