We will publish this transcript on our website, which is pdis.tw, for Public Digital Innovation Space.
Thank you very much, minister, for welcoming us. We were a larger group, but we’ve lost a few...
...along the way?
Yeah.
(laughter)
We’ve got the hard core here. Maybe we’ll just go around and introduce ourselves again. Vicki, would you?
Sure. Hi, Minister Tang. I’m Vicki. I’m from Uber. You’ve met some of my colleagues before. I’m very glad to meet with you again.
Right now we have been in Taiwan for about more than three years. We are also operating in a couple of core cities in Taiwan, but obviously facing some difficulties right now. We’d like to also seek some of your advice.
Sure.
Thank you.
My name’s Pongo Peng. I work for Gilead Science. As a bio pharma company we discover, develop, and commercialize drugs, particularly focused on the viral infectious disease, which is HIV, Hepatitis B, and Hepatitis C disease.
I’m David Katz from PayPal. We’re a fintech company, arguably the first at scale fintech company.
"The first fintech company."
I think here in Taiwan we really work with many thousands of merchants, mostly SMEs, to help them export goods and services globally by receiving payments from our platform. We help Taiwanese consumers be able to access goods and services globally, as well, by using their PayPal accounts to pay securely.
Perhaps later we could talk a little bit about there’s a lot more services that are a part of PayPal Inc. that you don’t necessarily think of when you think of PayPal. Maybe it’s something I can get into a little bit as you continue our dialog. I understand fintech is an area that Taiwan is looking to develop further. I’d be happy to talk about some of the things that we do that might work well in that context.
Again, I’m William Morin with Applied Materials. I’m also on the executive committee of US‑Taiwan Business Council. I don’t know if you’re familiar with Applied Materials.
Not terribly...
We are an equipment maker. We make the tools that are used to produce semi‑conductors and flat panel displays, as well as solar photovoltaic equipment. All of those are key industries here in Taiwan.
Taiwan’s actually our single biggest market. We manufacture our display tools down Tainan. We both sell here. We source here. We work with hundreds of local suppliers. We work with local universities. This is very important market for us.
It’s a pleasure to be here. We wanted to speak with you about it, to get your sense on what some of the new government programs in Asia Silicone Valley and so forth. Maybe just get a little progress report on that. Then we could also touch on individual issues. We come from different parts of the tech sector. We do have that common, we’re all technology based.
Hi. I’m Pearl, from Minister of Foreign Affairs.
Would you like to introduce us?
Hello. My name is Zach Huang. I’m assistant to Minister Tang. Here’s my name card.
(laughter)
How do we start?
Maybe if we could just...
Update you on the...
Yeah. I think that’d be really interesting to hear about the Asia Silicon Valley. I know that’s been in the works for probably not quite a year, six months or so. It would be great to understand...
Sure. It was part of the presidential campaign. Are you familiar with the redefinition story? Initially this used to be more of a regional development planning. As part of the president’s campaign, every large region in Taiwan has a distinct aspect to it, which is what we call the fthe original five, before the plus two. Now it’s five plus two plus two...
(laughter)
The original five were somewhat regional. Taoyuan hosts a lot of the production and distribution centers. Asia Silicon Valley is viewed as kind of unique because it proposes to build a test field for all kinds of new technologies, be it autonomous vehicles or smart logistics.
I’m sure that you know more about these fields and these actual operations than I do. [laughs] It was initially designed to be based mostly in the Taoyuan region. Even down through the areas like specific districts in Taoyuan, selected as our pilot-test sites. That was during the campaign. After the presidential election, there has been a redefinition.
It was then discovered that Taoyuan, this name of this very plan, 亞洲矽谷, in Mandarin it doesn’t really mean Asia and Silicon Valley. People looked at the name and said, "OK, this is 亞洲的矽谷 Asian Silicon Valley," like Taoyuan wants to be the Silicon Valley of Asia, which is unfortunate because it doesn’t really have to be this way.
People who took this impression, so they tie it into the speculation of land prices around Taoyuan, which to be fair is actually not that related to the Asia SV plan. But any inkling would be linked to this, "Oh, so Taoyuan wants to be the Silicon Valley of Asia," plan.
Then it was also discovered that part of this plan, especially around fintech, or around deregulation, or around virtual reality and augmented reality, it really isn’t region based. It’s more about creating a better startup system, about the fluidity of talents and the regulation harmonization. None of this is Taoyuan specific, so it doesn’t really make sense to base them on Taoyuan.
I participated in the initial discussion in the so‑called redefinition meeting and suggested that we put a dot between Asia and Silicon Valley, so that everybody knows that this plan is now 連結亞洲、連接矽谷 (linking Asia, connecting with Silicon Valley), instead of becoming a clone of Silicon Valley in Taoyuan, which is, to be honest, impossible.
The culture isn’t there, to name just one, which is also unfair to Taiwanese culture because we have a lot of unique cultures around here about thrift innovation, about a lot of consensus‑based policymaking, and things like that that weren’t found in Silicon Valley either. To bill it as the Asian Silicon Valley is to the detriment with both cultures.
It would be like the forced adaptation between two cultures, which nobody likes. That’s part of the definition. Now the Asia SV plan, as we know, has two main branches. One is the Internet of Things, which still needs a field to pass in, it still needs LoRa, IoT‑based spectrums, and other infrastructures.
It also has a startup promotion, and is a small culture. It’s about taking risks. It’s about making it easier to sell new commodities, about being able to securitize intentional assets, to name just a few. None of these are Taoyuan specific. These are the non‑client specific parts to start an ecosystem part.
While IoT, it’s not how you specify it, but it must be specific to a region. Taoyuan is just one of the first pilot sites. That was a redefinition. Our program so far is that I just received the Asia SV committee formation plan.
I’m one of the three ministers without portfolio to supervise this plan. The main leader is this independent agency, a company‑like thing that basically charts the entire investment plan. It has certain independence.
While we supervise this plan, it’s actually under the purview of the National Development Council, and also Deputy Minister Kung Ming‑hsin running this. Then he recruits a team of five to do the investment, and regulation, execution plan somewhat independent of the Cabinet.
As part of the Cabinet, there’s two other ministers without portfolio, Minister Chang Ching‑sen and me, who basically just look at those plans and make sure that it fits with the rest of the understanding of the other plans, like the desalination plan, the smart machine plan, or the other plans.
It’s our role in the Asia SV is just to make sure that they align with the other plans. For me, because my mandate, so to speak, my job is of the government which has transparency accountability in it. Another of my duty is to make sure that the ongoing spending of the Asia SV plan, which is a very large plan, is fiscally transparent.
Then all these projects that is carried by a ministry, the ministry must publish every one month and do this public platform called join.gov.tw, which, just like regulations.gov, announces every single regulation 60 days ahead of time so that people have a public commentary period starting this year.
Then it’s also a We the People petition platform, and now it’s also accountability platform for those presidential campaign‑promised spendings. It’s like three different US websites where I need to watch. That’s like the entry point for the open governance website at the moment.
Is that only for the executive branch or is the legislature will also participate in this join.gov.tw?
We’ve been talking with the legislative for well over a year now.
(laughter)
The original plan was that they would take the petition part first because it makes more sense. In UK, it actually started in the legislative and then gradually roll into city governments and administrative branch. It’s actually pretty peculiar in US that the administrative takes petitions before the legislative does.
In other parts of the world, it makes more sense for the MPs to take petitions because they translate to votes easy. [laughs] In any case, the MPs wants to regulate petitions. They have a different threshold. For administrative branch, we have 5,000. For legislative tier, I think they’re settling on 2,000.
The system needs some tweaking for it to work, but it will use the same website, but perhaps, when you get into their direction has this, "OK if you want to petition through your MPs, then click this," and then it would take you to LY, then join or something. It’s the same system and maintained by the same technical team.
I remember when a couple of these petitions as well around the people issue and I think it was a very good platform especially for the government agencies to respond to some of the questions. But I think, the one thing that we have observed is that a lot of the times, I don’t want to just focus on my issue.
How has the other government agencies have been responding to such a very innovative way of communicating with the public? Because I think, for our issues, it seems to be a very abrupt answer and I think some of the times people call for more specific answers. Have the agencies been receptive to this idea?
Yeah, sure. You see, for the Uber case, there’s actually two different cases. One is ministry initiated by the Ministry of Transport on the Taiwan who wants to talk about the Uber...
There was some...
But just the UberX case. But there’s a broader petition that says not just for UberX but for sharing economy in general.
It’s probably initiated by normal agencies, not by the government.
Not by the government. The petition was always started by a citizen, @nchild, who writes a lot of articles about sharing economy. You probably have heard of him, yeah?
Yeah.
That scope is much larger than the vTaiwan way. The vTaiwan just want to talk about the taxation, the insurance and the registration of UberX drivers. Nothing more nothing less. But the petition is asking for much more. It asks for a completely transparent process, do announcement and open negotiations. It asks for much more, but it has more than 5,000 people.
MOTC, I think, answer very eloquently but because the ask is so large, it’s actually more than what MOTC can handle. It fundamentally changes how the ministry has handled international negotiations. It’s asking for radical transparency on that.
It’s not something that MOTC can just say yes. MOTC says, "OK, this is what we have so far. The data you ask, we do have the data you ask. But if you ask about the production of such data, we can’t just decide ourselves that are there further negotiations and data," whatever. It must be open by default. It’s something beyond the MOTC’s mandate.
That’s before I went into a cabinet. I’ve been facilitating this kind of petition responses so that from now there will be much more substantial dialogue during the 60‑day period. In the MOTC case, the proposal waited for 60 days and then got some response but no communication between the day 1 and day 60, right?
I think there was just one response, yeah.
But for some other ministries, The Ministry of Health and Welfare, there’s a initial kickoff meeting about seven days after the proposal went through to make sure we know what the petition is asking for, and then a lot of kick‑off meetings but internally. There was outside experts and those were made public.
At least the transcript is made public and then on the 15 days, they release the infographics to explain the issue to general public. Then at 45 days, they had the initial statement and things like these. It’s every two weeks and on the third day, they announce the timetable, saying that we will do this every two weeks and so on.
We are now making this the norm and the waiting for 60 days the exception. That’s my work.
That’s great because I think a lot of the issues could be cross-ministerial as well and have your kind of coordination, it actually can answer questions as opposed to just pack into one...Exactly what you said, if they don’t have the scope and the authority to answer these questions, they...
Exactly. A ministry would answer for the one third which is their scope, but the other two third they have to ask somebody else. But to the proposer and the 5,000 people, because they imagine the administration as one...
Government is government.
[laughs] Yeah. That’s what we’re working very hard, to make it happen.
Interesting. Is there a model that you’re using? This feels quite cutting edge for government. Has any other country implemented anything like this? You mentioned there are pieces that you observed in US, like "We the People." This is an amalgamated approach and feels very integrated in a way that’s quite innovative.
Yeah, sure. We still look up to, like Iceland. Iceland is actually doing the much larger scale with the Pirate Party.
(laughter)
Of course.
Because that was their promise to do policymaking this way. Then, in individual cities like in Madrid or in Barcelona, they also do this very well because they also spread the budget‑making power, participatory budgeting, to ordinary citizens and in nearby countries in Uruguay and Portugal.
They’re now even entertaining the idea of voting for per se paper budgets using ATM, using automatic teller machines, so that even in rural areas people who can’t connect to their website, they could do this ATM stuff.
Essentially, if a cabinet or a government, being city or national level, it’s explicitly part of their campaign. They do this in a very integrated way. Otherwise, they tend do it with piece meal ‑‑ politically speaking. In Taiwan, President Tsai did make or the government part of her presidential promise. I think, I’m very lucky to do work for her, and for the nation to actually make those mandates.
That’s really interesting. Back on Asia Silicon Valley, what’s the time horizon for this? When do you expect to see result? You have metrics set out for results?
Sure. It’s all on the website. It’s in the ASV Development Agency. Have you seen the website?
No. I don’t know.
You said the inauguration in San Francisco, just last week when President Tsai was there.
Right.
Right. Yeah, the website.
Right. They were there ASVDA. Is it OK in Chinese?
For now, it’s fine. Sure.
Right. This is basically the outline of everything that they promised to do. It has four parts. It’s about a start of ecosystem. It’s about Innovative R&D Center that connects to Silicon Valley and with a national investment company that gets the things brought in. This is about field-testing IoT.
Finally, this is the regional part which is the demo, or the initial deregulation zone of these new autonomous speakers and things like these. As you can see, it still has the original thorium graphics on the right which is for regions and for logistics ‑‑ small logistics and smart medicine. Then, carriers and digital transformation of classical industrial areas respectively.
I think, this is pretty comprehensive obviously. For the first year, this planning center the ASVDA is on the planning stage. What it does is that it looks at automated streaks at all the regional governments, and find which existing projects align with these visions.
On the next budget cycle, it will then create new ministerial projects that didn’t exist but is necessary for this vision to happen. As well as advice the administration to maybe cut some budgets for the things that clearly doesn’t align. That’s going to be their project. They always say that here, as far as I’m concerned.
How is the agent in the US going to help with your effort in Taiwan? Because I think I’ve read something on President Tsai’s Facebook.
Yes, right. A lot of these C‑level people in the ASVDA came from the Silicon Valley. They were senior Silicon Valley entrepreneurs, researchers, scientists, or investors in the SV area.
The first year is about a cultural dissemination osmosis, if you would, [laughs] to take all the ministers and deputy ministers relevant to this project, and have them spend a lot of time with these C‑level people in ASVDA to understand this picture ‑‑ to just have a global picture. Nothing more, and nothing less.
They went there, they write for the next year’s yearly budget which is due August or something. They know what alignment means. That’s, I think, the super important thing. That’s what the overseer office does. They produce the recommendations and vital alignments, but also shifts people both ways so that people can gradually share the same horizon.
For Uber, some of the Uber mobility technology could probably contribute on some of these smart transportation as well. I would love to be taking part of some of that. Especially what you’re suggesting on the shank of ideas and learning about how some of this technology is being adopted. They are modern policy tools that we are interested to be helping.
That’s great. What I have said before was that during the planning stage — especially for ASVDA — what we, as ministers without portfolio, do is just to make sure they don’t conflict or clash with other important development programs. We have minimal input as of specifics like who do they visit, and things like that. That’s for the C‑level people in ASVDA to decide.
Got it. OK.
Interesting. It’s only focused on high tech, or also include biotech research and bio pharma?
Biotech may have its own ASVDA-like formulation. This is apparently trend-setting. If this extends, then we can have the medical or the bio sciences you said, and things like that, the green tech or whatever to follow this. For the medical and for bio things like that they’re also designing a very similar structure, but that’s still in the works.
This is an English website that I just sent you a link. Maybe we can try it?
In ASVDA?
OK. Got it.
But, I think, the translation has some shortcomings.
(laughter)
I think, this is still in the works. It’s not ready, which is why I was not showing it.
(laughter)
Yeah.
I think, there’s the preview. [laughs]
It hasn’t developed to understand fintech. It’s like we’re part of it. There is this financial technology experimentation bill in LY they’re looking at.
Yeah, the sandbox thing.
Is the sandbox connected to this, or do you see that as a separate initiative?
It is related, because it falls into deregulation, and also falls into the experimentation. It’s conceivable that a fintech experiment sandbox is not only limited by numbers of members or by the number of money but also regionally. It’s conceivably part of Asia SV.
Because Taiwan has this kind of legal system, Asia SV can only do what the law authorizes it does. For sandbox to happen we need LY to pass something that authorizes this experimentation...
...from the rules.
Exactly, yeah.
I think, the areas we are thinking about in the PayPal’s sync in terms of fintech development, obviously you know we’ve got the button on websites. Some of the areas that got increasingly involved in are developing the background engine for payments. What we call brain tree product which you may be familiar with.
That’s a lot of what will help make payments reduce the friction, and make it more seamless. A great example actually is Uber, where we power the payments underneath the Uber experience.
One of the things that I think people who come to Uber, I’ll use this as an example, that they find really satisfying as a customer is that experience where you order your car, you get in it. You take you ride, you get out, and you don’t stop to pay. It’s automatic. It happens. It’s seamless. People who get used to it then take a taxi at one point, and they try and walk out without paying because they get so used to that.
We see a lot of opportunity to reduce that friction to put in the background. A lot of other technologies that payments can be integrated in a way that could be more facilitate that development. Good examples would be contextual commerce, where you’re seeing that, people who are in websites.
It could be your Asia Silicon Valley Development Agency website where there’s a registration capability or whatever it is, and you can click on that. You’ve already had maybe a one‑click sign in where you’re integrated with that, where they can register and pay seamlessly.
Or, a better example is people who are looking at a picture on someone’s Facebook, and they like that vase. They say, "What is that?" They can literally click on the vase and buy it right there out of a photo. These are things that we see opportunities.
Another are voice set activator technologies. You’re seeing this increasingly. With artificial intelligence and ability to understand normal human speech, we now already see this integrated in their own. It’s like Lexus and echo devices in Siri, where you can,"Siri, buy me a coffee," or something. It’ll tell you where the coffee shop is, "Great, get me the double latte." It’ll make the order and the payment at the same time for you.
We’re seeing these kinds of technologies really starting to be ready. They’re not in development stage. I think, we’re going to see it more and more. The kinds of things like an Uber experience, you can see in other areas of commerce and, frankly, your life.
It’s an area where PayPal can provide expertise. We already have products in development around those types of technologies.
Just four months ago, I was still working with the Siri team. I was looking for this. [laughs] It really took a lot of development effort. Precisely as you said, this year is the year where people will take it for granted. I really do agree with everything you said.
Also, branching in on PayPal actually underlines a lot of inquisitive trust in the flow of commerce, because to pay for any service without even clicking, without even doing anything. Essentially, it’s line of trust.
Absolutely.
In this lining experience as like this, I think this is one of the fundamental parts that I do agree.
I think, it’s also the whole issue around trust of the data, both your payment credentials. It’s very important. What I would say is it’s also important then to have that artificial intelligence type machine learning where just because you’ve signed in, it’s still might not be you. You can have an account take over as that happens.
You need systems that are dynamic enough to say, "Wait a second. RT doesn’t behave this way. That’s not right. I’m going to stop this." It creates some friction, but it’s for a good reason because instead of the normal place where you might pick up your latte or whatever it is, suddenly you’re across town and you’re trying to buy sports equipment. That just doesn’t match what your behavior is.
I think, having that dynamic capability is very important, I’d add. There’s also a lot of development in the PDP space where I think that’s an opportunity also, as you think about the sandbox. We see this in the United States certainly.
One of our most popular products in the area is a product called Venmo, which last year or the year before was the third most downloaded app in the US. It’s incredibly popular with millennials. I’m sure there are products like this developing here in your country, but it’s something that we think we got a lot of expertise in helping make movement of cash cashless.
Digital movement between social groups, friends, etc. It can be delivered, in that case, at scale at no cost to the consumer. There are other ways to monetize it later, but again, these are things that we can see as being part of the broadening ecosystem around fintech here in Taiwan.
The error correction example that you just mentioned, I think this year is the year where machines reliably does this better than humans. It used to be a unique, best actor to be sure, but now... [laughs]
The flip side of this is, of course, what we call regulatory technology. The regulators who used to outperform machines in telling one bad actor over a good actor, or whatever, inside the sandbox.
Now, they are faced with very diverse biometric data that also needs machine learning to be even able to regulate well. Not to tell whether this should go out of sandbox to be one of the original sellers, or to graduate but failing one or two criteria, so they have to go back and fill in those criteria.
Those decisions, you can’t build this on human alone now, because of the sheer diversity and the volume of the data.
Too much data for humans...
It’s too much data, and the error rate, even 0.3 percent or something, is very significant from machine view point. If you spot check, it’s not going to be spot checked.
I agree.
That’s also what we are trying to develop, and of course, any help will be very appreciated.
An area where Taiwan can play a really leading role, there’s a lot of talk around KYC ‑‑ know your customer. Pretty much all regulators around the world, at some point fall back to what is fundamentally like a 19th century technology. Show up at a physical place, show a document that’s your photo on whatever ID you’ve got, and they stamp it and say, "OK. You’re who you say you are," and they are done. They’ve done their KYC.
What our view is, I would challenge everyone in this room, that if I ask you what says more about your identity, your one physical ID that you could pull out ‑‑ I’m sure you got one ‑‑ or your mobile, where that mobile is, your behavior, what you’re buying.
We think that there’s an opportunity to have a much more dynamic approach to identifying someone that doesn’t run on one touch point using one ID, but can really, in a dynamic way, to the example I gave earlier and say, "Wait a second. That’s not you. That’s just not just who you are," and do the due diligence.
We think that a better way to secure the system and stop people from fraud, and bad behaviors like funding terrorists. I think Taiwan can really play a leading role through the sandbox approach, and through the Silicon Valley, doing experiments. How do you get the right balance?
It won’t be perfect, it won’t be, but then again, humans definitely are not perfect. We know someone showing up with an ID at a window, there’s an error right there, and it’s probably pretty high, because fake IDs are pretty easy to get.
Whereas having a dynamic approach, I think we can get a much better resolve for what we want to get from regulatory perspective, and it’s going to be a better experience for consumers and merchants.
Three or four years ago when Taiwan was working on big data and so on, the state of the art was still that you needed volumes of tagged data for supervised learning, and things like that. You need a lot of data format in the same way, and maybe...
Yeah. Structured data.
...two or three data sources as primary data sources, and have to be authenticated to be minimally noise free. This year and end of last year, we had a series of breakthroughs in AI, including CNN meta-modeling. Now, what you want is a lot of ambient data.
They each can contain a lot of noise, even user generated, it doesn’t care. Then the machine learning algorithm can work with others and can get a much more clear picture than what it used to be. It certainly less expensive, because you don’t have to be structured all the way. We are very interested in developing this.
How do you balance all that privacy concerns? Is that in your portfolio as well?
In the cyber security part, like against malicious attackers, that’s the cyber security office in the administration. Director Howard Jyan is heading that, and because my time I’ve worked here, is that everything I can see and hear is public information, it’s FOIA-compatible. I’m not in touch with that, because it’s mostly national secret, [laughs] and confidential information.
There is a dedicated office for malicious attacks and prevention against those creating infrastructures, but I’m not at all in touch with them. I design. Your privacy is not in any threat by malicious attackers, it’s also by benevolent corporations who just didn’t calculate their risk of linked data, of PIIs with other operators.
That is much harder to quantify. Academically, there’s ways to quantify, but I have yet to see any regulatory environment that really takes differential privacy, or other quantifiable algorithms into part of the rule of law. If we crack that, it would be the world’s best, but to be very honest, it’s not likely to happen soon. [laughs]
This is like algorithmic regulation of privacy, it’s something as a researcher I’m very interested in, but realistically, I don’t think that’s going to happen this year or the next. What we are doing is that we are just translating and facilitating the understanding of those ISO standards to 9100, 101 for due diligence, to minimize, but not eliminate the risk.
Make sure that all the, especially open data, that other ministries think twice, and report to their internal DPA inside the ministry, or their deputy ministry, and consult with external experts, which must be at least one third of the open data port of each ministry.
Consult with at least with those data scientists and customer advocates or academics, before thinking, "Yeah, it’s not OK to release." This is not a perfect setup because it relies on human judgment, and it doesn’t have that much of cross ministerial correlation.
It’s possible that the release of the data, or APIs, that passes the review of those boards of those ministries. But taken together, it is actually a privacy concern that nobody managed to catch. I worry about it a lot, but there’s no realistic way to fix it this year.
I think we’ve observed that in many cases, the consumer consent, or the user consenting for the data to be used in particular ways, or you just using metadata, so it can’t be ascribed to an individual, are two of the approaches.
What’s very interesting as we get into IoT, I don’t know if you saw this news that there’s a case in the United States ‑‑ a murder case ‑‑ where they are seeking to submit Alexa as a witness, the echo machine, because it may have recorded relevant voice, relevant sound pertaining to the case.
I can imagine this would be a fascinating area of law, and frankly, how this as more and more devices are enabled and are recording. You have a smart TV that’s ready for voice activated remote, it might be always on standby, ready to hear what you say, who knows? At some point, what does that mean for us? It will be really interesting to see how this case plays out.
It would. We will be one experimentation side out of many of this open government consensus‑making mechanism, to assess what the society feels comfortable with. That’s probably the best we can do for the moment.
I want to echo David’s earlier point about his example on KYC. I just thinking, because all we know a lot in terms of the regulatory environment in Taiwan. There’s a lot of licenses and a lot of restrictions with that.
I’m sure for our case, I think one interesting thing is that we are developing telematics technology, where we are able to have dynamic feedback about drivers’ behavior, which could potentially be more credible than your preferential driver’s license that you’ve applied 10 years ago.
Are you respective to these new ways to looking into where to get the credibility of a certain person’s skill sets? How are you seeing that as a relationship with the current regulatory environment in Taiwan?
I did not work with the Apple car team [laughs]. It was one of the motivations for sure, because then, it’s just like your family. It’s a symbiotic relationship with you, and it becomes an extension of you, by it I mean the car. As a rule maker, this is very difficult to reconcile with any kind of regulatory mechanism that’s based on supervision.
If you have experimental sandbox, where you say, "We don’t regulate anything as long as you do what you claim in your business plan, but it has to be risk controlled," then this, we can do.
To scale this out to say, "By default, we are going to regulate, and if something happens, we go back and punish the people who caused this," it’s politically very not very feasible to introduce this kind of wide range trial first way, which is part of the beauty of the Asia SV plan, is that the region can be arbitrarily large. It can be Taiwan if you manage to convince the consensus of the Taiwan people.
As a living lab.
As a living lab. [laughs] They always start small. If you can get a wild, remote area working, and it actually solves the demands of the remote area, but that remote area being an isolated sandbox, then you can scale it out to nearby townships and whatnot. It never works the other way around.
There are some insurance companies where consumers can chose consent to have a monitoring device in their vehicle, and then that affects their insurance rate. If they are seen as a good driver, better rate, bad driver, worse rate.
That’s a little of what we were talking about before, this is an informed consent, yes, you can see how well I’m driving over a period of time.
In that case, it’s the private sector serving as a test bed. Their regulators, if this is proven to work can say, "OK. Now, we work with the private sector then to make it part of the regulatory framework." It’s always tested and work somewhere, be it a private sector or a civil society.
It seems like there are a lot of opportunities with the Asian Silicon Valley ASVDA?
Yes.
.
I think it would be great to know as you are working out the plans of how some of executional stuff that we could participate, I think would be very interesting, because I think right now it’s like living in utopia. You have all these great concepts of things that should be done, and when it comes to the execution phase, I think it will be very interesting to see.
Even this year, we have a lot of infrastructure work. Regionally, there is the lower IoT communication architecture. If you have road and the high speed rails, the 4G connection doesn’t even roam.
(laughter)
Also, the basic mapping data from each regional government, versus the national government and Ministry of Interior’s map, all these angles need to be in place before any of those utopian projects can run. Otherwise, if you are running through the border of a county and the car stops, what...?
(laughter)
This year, we still have a lot to do as part of the digital nation plan. It is a separate plan to reconcile what the regional governments has in its data. That is regional population and environment, and reconcile it with the national database. We are still working on that.
The regional government, are they willing? Are they eager partners, or is this...? It is new, and oftentimes there’s resistance to the new. I am just curious, what happens?
It’s new in the sense that it’s a new thing that is starting with next year’s budget. We are now trying a way to make digital infrastructure part of the infrastructure budget. This has never happened before. We still need LY’s approval. If this works, then it’s just like railway or anything.
Obviously, we just don’t think the governments would like it because it’s not MIS anymore. It’s not just management of information systems. That’s the idea.
It’s all very exciting.
Is Taiwan feeling a little bit left out, because they were supposed to be before you defined them that defined a project, they had the expectation that they were going to be the... [laughs]
To be the early place where there autonomous cars?
Yeah.
(laughter)
Mayor Cheng is very open minded. I’m sure that he’s not opposed to other places like Kaohsiung. Kaohsiung is doing a lot, actually, autonomous vehicles. I think that this October they are going to have one of the autonomous vehicles, in a point‑to‑point fixed route, but still starting to run.
By opening up the pilot sites to selected other cities instead of Taiwan, that also gives Taiwan a lot of...Because frankly speaking, Taiwan’s IT budget before the Asia SV plan does not compare to Taipei’s of Kaohsiung’s.
If Taipei, or Kaohsiung, or Taitung, or Tainan has something that’s already proven using their existing IT budget, then it’s better for Taiwan, because then they can scale it, being a relatively less populace area, or things like that. I’m sure it’s pretty big for them also.
You’re really groundbreaking here. Any other points or comments to make? You’ve been very generous with your time.
It’s fine.
It’s really interesting, and I obviously wish you the best. There are competitors out there as you know. I’m based in Singapore, they have a smart‑based initiative. I don’t know if you, within office under the PM, led by the Foreign Minister. I’m sure you’ve seen their plans.
They are also quite ambitious, but you are right up there in your ambition. That’s great to see. Competition is great for everyone.
Our office PD is just like Singapore’s GDS and so on. We are based on open access and open source. Frankly speaking, we use each other’s focus all the time.
Absolutely.
It’s not a competition.
It’s not a zero-sum at all.
That’s right...
That’s a paradigm. I’m sure you’ll learn lessons from them, they’ll learn lessons from you, and both will be better for it.
I just have one last question, we have more questions here, but in terms of a lot of the new businesses, a lot of the times the business model across different agencies, and different areas of issues.
One thing that we are not just asked by a couple of the startup companies that we know is that it’s always difficult to find one contact window, or one person within the government to look at the issues as a whole.
Is there something on your agenda? Other than the join.gov.tw, is there something you’re working on as well, to make it easier for...?
Definitely.
...to communicate?
It depends on what exactly is that you are asking. If it’s a regulatory question, then part of the innovation plan, the sandbox plan, it would be the regulatory attack part of it, which as part of the definition, answers these kind of questions.
It’s only pertaining to fintech at the moment, because it’s drafted by the FSC. We did talk about scaling this to other parts of the ministries, perhaps copying and pasting the same language.
(laughter)
It’s just talking at the moment, because nobody really knows if FSC, or traditional regulatory ministry can just overnight becoming a RegTech developer. We design what we can into that act, and provide them with sufficient resource.
As you see, there is no international predecessors. This is still new. Sandbox was two years ago in UK or something. Nobody knows if this would work. If this works, then I’m sure the other ministries would look up to it as an example.
Even if this fails, interestingly, we would have a postmortem, [laughs] so that other ministries avoid these mistakes while we provide the revision of the act. We did that with crowdfunding with other regulations just by adjusting experimenters, until the market and the civil society is OK with it.
For some trials like the Coarsely Health Corporation, we got it somewhat right in the first try. Now, we are looking to scale it out. It’s all very experimental, but we are looking into scaling it out.
There’s a role for international companies involved. By all means, I hope you’ll find a way to let us know, so that we can sort out how we can help.
Sure.
We are from the US‑Taiwan Business Council, and as a group, as you were saying as well, a lot of technology companies, so both individual and even as a group, we love to explore how we can support the initiative.
Absolutely. Silicon Valley, I don’t think it would be possible without Taiwan. If you look at all the human capital, and the back and forth. We would like to be part of this as well, just as Taiwan has been part of Silicon Valley in California. It’s all very exciting.
Thank you.
Great.
Thank you very much.
Thank you.
Thank you very much.
Thank you very much for staying late for us.
It’s fine.