English subtitles

← How digital innovation can fight pandemics and strengthen democracy

Get Embed Code
25 Languages

Showing Revision 11 created 07/02/2020 by Erin Gregory.

  1. Audrey Tang: Very happy to be joining you,
  2. and good local time, everyone.
  3. David Biello: So, tell us about --

  4. Sorry to --
  5. Tell us about digital tools and COVID.
  6. AT: Sure.

  7. Yeah, I'm really happy to share with you
  8. how Taiwan successfully
    countered the COVID
  9. using the power
    of digital democracy tools.
  10. As we know, democracy improves
    as more people participate.
  11. And digital technology remains one
    of the best ways to improve participation,
  12. as long as the focus
    is on finding common ground,
  13. that is to say, prosocial media
    instead of antisocial media.
  14. And there's three key ideas
    that I would like to share today
  15. about digital democracy
    that is fast, fair and fun.
  16. First about the fast part.

  17. Whereas many jurisdictions began
    countering coronavirus only this year,
  18. Taiwan started last year.
  19. Last December, when Dr. Li Wenliang,
    the PRC whistleblower,
  20. posted that there are new SARS cases,
  21. he got inquiries
    and eventually punishments
  22. from PRC police institutions.
  23. But at the same time,
  24. the Taiwan equivalent
    of Reddit, the Ptt board,
  25. has someone called nomorepipe
  26. reposting Dr. Li Wenliang's
    whistleblowing.
  27. And our medical officers
    immediately noticed this post
  28. and issued an order that says
  29. all passengers flying in
    from Wuhan to Taiwan
  30. need to start health inspections
    the very next day,
  31. which is the first day of January.
  32. And this says to me two things.

  33. First, the civil society
    trusts the government enough
  34. to talk about possible
    new SARS outbreaks in the public forum.
  35. And the government trusts citizens enough
  36. to take it seriously and treat it
    as if SARS has happened again,
  37. something we've always
    been preparing for, since 2003.
  38. And because of this open civil society,
  39. according to the CIVICUS Monitor
    after the Sunflower Occupy,
  40. Taiwan is now the most open society
    in the whole of Asia.
  41. We enjoy the same freedom
    of speech, of assembly,
  42. [unclear] as other liberal democracies,
  43. but with the emphasis
    on keeping an open mind
  44. to novel ideas from the society.
  45. And that is why our schools
    and businesses still remain open today,
  46. there was no lockdown,
  47. it's been a month
    with no local confirmed cases.
  48. So the fast part.

  49. Every day, our Central Epidemic
    Command Center, or CECC,
  50. holds a press conference,
    which is always livestreamed,
  51. and we work with the journalists,
  52. they answer all the questions
    from the journalists,
  53. and whenever there's a new idea
    coming in from the social sector,
  54. anyone can pick up
    their phone and call 1922
  55. and tell that idea to the CECC.
  56. For example, there was one day in April

  57. where a young boy has said
    he doesn't want to go to school
  58. because his school mates may laugh at him
  59. because all he had is a pink medical mask.
  60. The very next day,
  61. everybody in the CECC press conference
    started wearing pink medical masks,
  62. making sure that everybody learns
    about gender mainstreaming.
  63. And so this kind of rapid response system
  64. builds trust between the government
    and the civil society.
  65. And the second focus is fairness.

  66. Making sure everybody can use
    their national health insurance card
  67. to collect masks from nearby pharmacies,
  68. not only do we publish the stock level
    of masks of all pharmacies,
  69. 6,000 of them,
  70. we publish it every 30 seconds.
  71. That's why our civic hackers,
    our civil engineers in the digital space,
  72. built more than 100 tools
    that enable people to view a map,
  73. or people with blindness
    who talk to chat bots, voice assistants,
  74. all of them can get the same
    inclusive access to information
  75. about which pharmacies near them
    still have masks.
  76. And because the national
    health insurance single payer
  77. is more than 99.9 percent
    of health coverage,
  78. people who show any symptoms
  79. will then be able to take
    the medical mask,
  80. go to a local clinic,
  81. knowing fully that they will
    get treated fairly
  82. without incurring any financial burden.
  83. And so people designed a dashboard
  84. that lets everybody see
    our supply is indeed growing,
  85. and whether there's over- or undersupply,
  86. so that we codesign
    this distribution system
  87. with the pharmacies,
    with the whole of society.
  88. So based on this analysis,

  89. we show that there was
    a peak at 70 percent,
  90. and that remaining 20 percent of people
    were often young, work very long hours,
  91. when they go off work,
    the pharmacies also went off work,
  92. and so we work with convenience stores
  93. so that everybody can collect
    their mask anytime,
  94. 24 hours a day.
  95. So we ensure fairness of all kinds,
  96. based on the digital democracy's feedback.
  97. And finally, I would like to acknowledge
    that this is a very stressful time.

  98. People feel anxious, outraged,
  99. there's a lot of panic buying,
  100. a lot of conspiracy theories
    in all economies.
  101. And in Taiwan,
  102. our counter-disinformation
    strategy is very simple.
  103. It's called "humor over rumor."
  104. So when there was a panic buying
    of tissue paper, for example,
  105. there was a rumor that says,
  106. "Oh, we're ramping up mass production,
  107. it's the same material as tissue papers,
  108. and so we'll run out
    of tissue paper soon."
  109. And our premier showed
    a very memetic picture
  110. that I simply have to share with you.
  111. In very large print,
  112. he shows his bottom,
  113. wiggling it a little bit,
  114. and then the large print says
  115. "Each of us only have
    one pair of buttocks."
  116. And of course, the serious table shows
  117. that tissue paper came
    from South American materials,
  118. and medical masks
    come from domestic materials,
  119. and there's no way that ramping up
    production of one
  120. will hurt the production of the other.
  121. And so that went absolutely viral.
  122. And because of that,
    the panic buying died down
  123. in a day or two.
  124. And finally, we found out the person
    who spread the rumor in the first place
  125. was the tissue paper reseller.
  126. And this is not just
    a single shock point in social media.

  127. Every single day,
  128. the daily press conference gets translated
  129. by the spokesdog of the Ministry
    of Health and Welfare,
  130. that translated a lot of things.
  131. For example, our physical distancing
    is phrased as saying
  132. "If you are outdoors,
    you need to keep two dog-lengths away,
  133. if you are indoor,
    three dog-lengths away," and so on.
  134. And hand sanitation rules, and so on.
  135. So because all this goes viral,
  136. we make sure that the factual humor
    spreads faster than rumor.
  137. And they serve as a vaccine,
    as inoculation,
  138. so that when people see
    the conspiracy theories,
  139. the R0 value of that will be below one,
  140. meaning that those ideas will not spread.
  141. And so I only have
    this five-minute briefing,

  142. the rest of it will be driven
    by your Q and A,
  143. but please feel free to read more
  144. about Taiwan's
    counter-coronavirus strategy,
  145. at taiwancanhelp.us.
  146. Thank you.
  147. DB: That's incredible.

  148. And I love this "humor versus rumor."
  149. The problem here in the US, perhaps,
  150. is that the rumors seem to travel
    faster than any response,
  151. whether humorous or not.
  152. How do you defeat that aspect in Taiwan?
  153. AT: Yeah, we found that, of course,

  154. humor implicitly means
    there is a sublimation
  155. of upsetness, of outrage.
  156. And so as you see, for example,
    in our premier's example,
  157. he makes fun of himself.
  158. He doesn't make a joke
    at the expense of other people.
  159. And this was the key.
  160. Because people think it hilarious,
  161. they share it,
  162. but with no malicious or toxic intentions.
  163. People remember the actual payload,
  164. that table about materials
    used to produce masks,
  165. much more easily.
  166. If they make a joke
    that excludes parts of the society,
  167. of course, that part of society
    will feel outraged
  168. and we will end up
    creating more divisiveness,
  169. rather than prosocial behavior.
  170. So the humor at no expense,
  171. not excluding any part of society,
  172. I think that was the key.
  173. DB: It's also incredible

  174. because Taiwan has such close ties
    to the origin point of this.
  175. AT: PRC, yes.

  176. DB: The mainland.

  177. So given those close economic ties,
  178. how do you survive
    that kind of disruption?
  179. AT: Yeah, I mean, at this moment,

  180. it's been almost a month now
    with no local confirmed cases,
  181. so we're doing fine.
  182. And what we are doing, essentially,
  183. is just to respond faster
    than pretty much anyone.
  184. We started responding last year,
  185. whereas pretty much everybody else
    started responding this year.
  186. We tried to warn the world
    last year, but, anyway.
  187. So in any case,
  188. the point here is
    that if you start early enough,
  189. you get to make sure
    that the border control
  190. is the main point where you quarantine
    all the returning residents and so on,
  191. instead of waiting until
    the community spread stage,
  192. where even more human-right
    invading techniques
  193. would probably have to be deployed
    one way or the other.
  194. And so in Taiwan, we've not declared
    an emergency situation.

  195. We're firmly under the constitutional law.
  196. Because of that, every measure
    the administration is taking
  197. is also applicable
    in non-coronavirus times.
  198. And this forces us to innovate.
  199. Much as the idea of
    "we are an open liberal democracy"
  200. prevented us from doing takedowns.
  201. And therefore, we have to innovate
    of humor versus rumor,
  202. because the easy path,
    the takedown of online speech,
  203. is not accessible to us.
  204. Our design criteria,
    which is no lockdowns,

  205. also prevented us
    from doing any, you know,
  206. very invasive privacy encroaching
    response system.
  207. So we have to innovate at the border,
  208. and make sure that we have
    a sufficient number of, for example,
  209. quarantine hotels
    or the so-called "digital fences,"
  210. where your phone is basically connected
    to the nearby telecoms,
  211. and they make sure that if they go out
    of the 15-meter or so radius,
  212. an SMS is sent to the local
    household managers or police and so on.
  213. But because we focus
    all these measures at the border,
  214. the vast majority of people
    live a normal life.
  215. DB: Let's talk about that a little bit.

  216. So walk me through the digital tools
  217. and how they were applied to COVID.
  218. AT: Yes.

  219. So there's three parts
    that I just outlined.
  220. The first one is the collective
    intelligence system.
  221. Through online spaces
  222. that we design to be devoid
    of Reply buttons,
  223. because we see that,
    when there's Reply buttons,
  224. people focus on each other's
    face part, not the book part,
  225. and without "Reply" buttons,
  226. you can get collective intelligence
  227. working out their rough consensus
    of where the direction is going
  228. with the response strategies.
  229. So we use a lot of new technologies,
  230. such as Polis,
  231. which is essentially a forum
    that lets you upvote and downvote
  232. each other's feelings,
  233. but with real-time clustering,
  234. so that if you go to cohack.tw,
  235. you see six such conversations,
  236. talking about how to protect
    the most vulnerable people,
  237. how to make a smooth transition,
  238. how to make a fair
    distribution of supplies and so on.
  239. And people are free to voice their ideas,

  240. and upvote and downvote
    each other's ideas.
  241. But the trick is that we show people
    the main divisive points,
  242. and the main consensual points,
  243. and we respond only to the ideas
  244. that can convince
    all the different opinion groups.
  245. So people are encouraged
    to post more eclectic, more nuanced ideas
  246. and they discover,
    at the end of this consultation,
  247. that everybody, actually,
    agrees with most things,
  248. with most of their neighbors
    on most of the issues.
  249. And that is what we call
    the social mandate,
  250. or the democratic mandate,
  251. that then informs our development
    of the counter-coronavirus strategy
  252. and helping the world with such tools.
  253. And so this is the first part,

  254. it's called listening at scale
    for rough consensus.
  255. The second part I already covered
    is the distribute ledger,
  256. where everybody can go
    to a nearby pharmacy,
  257. present their NHI card,
    buy nine masks, or 10 if you're a child,
  258. and see the stock level
    of that pharmacy on their phone
  259. actually decreasing by nine or 10
    in a couple of minutes.
  260. And if they grow by nine or 10,
  261. of course, you call the 1922,
  262. and report something fishy is going on.
  263. But this is participatory accountability.
  264. This is published every 30 seconds.
  265. So everybody holds each other accountable,
  266. and that massively increases trust.
  267. And finally, the third one,
    the humor versus rumor,

  268. I think the important thing to see here
  269. is that wherever there's a trending
    disinformation or conspiracy theory,
  270. you respond to it with a humorous package
  271. within two hours.
  272. We have discovered,
    if we respond within two hours,
  273. then more people see the vaccination
    than the conspiracy theory.
  274. But if you respond four hours
    or a day afterwards,
  275. then that's a lost cause.
  276. You can't really counter that
    using humor anymore,
  277. you have to invite the person
    who spread those messages
  278. into cocreation workshops.
  279. But we're OK with that, too.
  280. DB: Your speed is incredible.

  281. I see Whitney has joined us
    with some questions.
  282. Whitney Pennington Rodgers: That's right,

  283. we have a few coming in already
    from the audience.
  284. Hi there, Audrey.
  285. And we'll start with one
    from our community member Michael Backes.
  286. He asks how long has humor
    versus rumor been a strategy
  287. that you've implemented.
  288. Excuse me.
  289. "How long has humor versus rumor
    strategy been implemented?
  290. Were comedians consulted
    to make the humor?"
  291. AT: Yes, definitely.

  292. Comedians are our most
    cherished colleagues.
  293. And each and every ministry has a team
    of what we call participation officers
  294. in charge of engaging
    with trending topics.
  295. And it's a more than 100
    people-strong team now.
  296. We meet every month
    and also every couple of weeks
  297. on specific topics.
  298. It's been like that since late 2016,
  299. but it's not until our previous
    spokesperson, Kolas Yotaka,
  300. joined about a year and a half ago,
  301. do the professional comedians
    get to the team.
  302. Previously, this was more about inviting
    the people who post, you know,
  303. quotes like "Our tax filing system
    is explosively hostile,"
  304. and gets trending,
  305. and previously, the POs
    just invited those people.
  306. Everybody who complains
  307. about the finance minister's
    tax-filing experience
  308. gets invited to the cocreation
    of that tax filing experience.
  309. So previously, it was that.
  310. But Kolas Yotaka and the premier
    Su Tseng-chang said,
  311. wouldn't it be much better
    and reach more people
  312. if we add some dogs to it
    or cat's pictures to it?
  313. And that's been around
    for a year and a half.
  314. WPR: Definitely, I think it makes
    a lot of difference, just even seeing them

  315. without being part
    of the thought process behind that.
  316. And we have another question here
    from G. Ryan Ansin.
  317. He asks, "What would you rank
    the level of trust
  318. your community had before the pandemic,
  319. in order for the government
    to have a chance
  320. at properly controlling this crisis?"
  321. AT: I would say that a community
    trusts each other.

  322. And that is the main point
    of digital democracy.
  323. This is not about people
    trusting the government more.
  324. This is about the government
    trusting the citizens more,
  325. making the state transparent
    to the citizen,
  326. not the citizen transparent to the state,
  327. which would be some other regime.
  328. So making the state
    transparent to the citizens
  329. doesn't always elicit more trust,
  330. because you may see something wrong,
    something missing,
  331. something exclusively hostile
    to its user experience,
  332. an so on, of the state.
  333. So it doesn't necessarily lead
    to more trust from the government.
  334. Sorry, from the citizen to the government.
  335. But it always leads to more trust
    between the social sector stakeholders.
  336. So I would say the level of trust
    between the people

  337. who are working on, for example,
  338. medical officers,
  339. and people who are working
    with the pandemic responses,
  340. people who manufacture medical masks,
  341. and so on,
  342. all these people,
  343. the trust level between them is very high.
  344. And not necessarily
    they trust the government.
  345. But we don't need that
    for a successful response.
  346. If you ask a random person on the street,
  347. they will say Taiwan is performing so well
    because of the people.
  348. When the CECC tells us to wear the mask,
  349. we wear the mask.
  350. When the CECC tells us not to wear a mask,
  351. like, if you are keeping
    physical distance,
  352. we wear a mask anyway.
  353. And so because of that,
  354. I think it's the social sector's trust
    between those different stakeholders
  355. that's the key to the response.
  356. WPR: I will come back shortly
    with more questions,

  357. but I'll leave you guys
    to continue your conversation.
  358. AT: Awesome.

  359. DB: Well, clearly,
    part of that trust in government

  360. was maybe not there in 2014
    during the Sunflower Movement.
  361. So talk to me about that
  362. and how that led to this,
    kind of, digital transformation.
  363. AT: Indeed.

  364. Before March 2014, if you asked
    a random person on the street in Taiwan,
  365. like, whether it's possible
    for a minister -- that's me --
  366. to have their office in a park,
    literally a park,
  367. anyone can walk in and talk to me
    for 40 minutes at a time,
  368. I'm currently in that park,
    the Social Innovation Lab,
  369. they would say that this is crazy, right?
  370. No public officials work like that.
  371. But that was because on March 18, 2014,
  372. hundreds of young activists,
    most of them college students,
  373. occupied the legislature
  374. to express their profound opposition
    to a trade pact with Beijing
  375. under consideration,
  376. and the secretive manner in which
    it was pushed through the parliament
  377. by Kuomintang,
    the ruling party at the time.
  378. And so the protesters
    demanded, very simply,
  379. that the pact be scraped,
  380. and the government to institute
    a more transparent ratification process.
  381. And that drew widespread public support.
  382. It ended a little more
    than three weeks later,

  383. after the government promised and agreed
  384. on the four demands [unclear]
    of legislative oversight.
  385. A poll released after the occupation
  386. showed that more than 75 percent
    remained dissatisfied
  387. with the ruling government,
  388. illustrating the crisis of trust
    that was caused by a trade deal dispute.
  389. And to heal this rift
    and communicate better
  390. with everyday citizens,
  391. the administration reached out
    to the people who supported the occupiers,
  392. for example, the g0v community,
  393. which has been seeking
    to improve government transparency
  394. through the creation of open-source tools.
  395. And so, Jaclyn Tsai,
    a government minister at the time,
  396. attended our hackathon
  397. and proposed the establishment
    of novel platforms
  398. with the online community
    to exchange policy ideas.
  399. And an experiment was born called vTaiwan,

  400. that pioneerly used tools such as Polis,
  401. that allows for "agree" or "disagree"
    with no Reply button,
  402. that gets people's rough consensus
    on issues such as crowdfunding,
  403. equity-based crowdfunding, to be precise,
  404. teleworking and many other
    cyber-related legislation,
  405. of which there is no existing
    unions or associations.
  406. And it proved to be very successful.
  407. They solved the Uber problem, for example,
  408. and by now, you can call an Uber --
  409. I just called an Uber this week --
  410. but in any case,
    they are operating as taxis.
  411. They set up a local
    taxi company called Q Taxi,
  412. and that was because on the platform,
    people cared about insurance,
  413. they care about registration,
  414. they care about all the sort of,
    protection of the passengers, and so on.
  415. So we changed the taxi regulations,
  416. and now Uber is just another taxi company
  417. along with the other co-ops.
  418. DB: So you're actually, in a way,

  419. crowdsourcing laws
    that, well, then become laws.
  420. AT: Yeah, learn more at crowd.law.

  421. It's a real website.
  422. DB: So, some might say
    that this seems easier,

  423. because Taiwan is an island,
  424. that maybe helps you control COVID,
  425. helps promote social cohesion,
  426. maybe it's a smaller country than some.
  427. Do you think that this could be
    scaled beyond Taiwan?
  428. AT: Well, first of all,

  429. 23 million people
    is still quite some people.
  430. It's not a city,
  431. as some usually say, you know,
    "Taiwan is a city-state."
  432. Well, 23 million people,
    not quite a city-state.
  433. And what I'm trying to get at,
  434. is that the high population density
    and a variety of cultures --
  435. we have more than 20 national languages --
  436. doesn't necessarily lead
    to social cohesion, as you said.
  437. Rather, I think, this is the humbleness
    of all the ministers
  438. in the counter-coronavirus response.
  439. They all took on an attitude
    of "So we learned about SARS" --
  440. many of them were in charge
    of the SARS back then,
  441. but that was classical epidemiology.
  442. This is SARS 2.0,
    it has different characteristics.
  443. And the tools that we use
    are very different,
  444. because of the digital transformation.
  445. And so we are in it to learn
    together with the citizens.
  446. Our vice president at the time,

  447. Dr. Chen Chien-jen, an academician,
  448. literally wrote the textbook
    on epidemiology.
  449. However, he still says,
  450. "You know, what I'm going to do
    is record an online MOOC,
  451. a crash course on epidemiology,
  452. that shares with,
  453. I think, more than 20,00 people
    enrolled the first day,
  454. I was among them,
  455. to learn about important ideas,
  456. like the R0 and the basic transmission
  457. and how the various
    different measures work,
  458. and then they asked people to innovate.
  459. If you think of a new way
    that the vice president did not think of,
  460. just call 1922,
  461. and your idea will become
    the next day's press conference.
  462. And this is this colearning strategy,

  463. I think, that more than anything
    enabled the social cohesion,
  464. as you speak.
  465. But this is more of a robust
    civil society than the uniformity.
  466. There's no uniformity at all in Taiwan,
  467. everybody is entitled to their ideas,
  468. and all the social innovations,
  469. ranging from using
    a traditional rice cooker
  470. to revitalize, to disinfect the mask,
  471. to pink medical mask, and so on,
  472. there's all variety
    of very interesting ideas
  473. that get amplified
    by the daily press conference.
  474. DB: That's beautiful.

  475. Now -- oh, Whitney is back,
  476. so I will let her ask the next question.
  477. WPR: Sure, we're having
    some more questions come in.

  478. One from our community member Aria Bendix.
  479. Aria asked, "How do you ensure
    that digital campaigns act quickly
  480. without sacrificing accuracy?
  481. In the US, there was a fear
    of inciting panic about COVID-19
  482. in early January."
  483. AT: This is a great question.

  484. So most of the scientific ideas
    about the COVID are evolving, right?
  485. The efficacy of masks, for example,
    is a very good example,
  486. because the different characteristics
    of previous respiratory diseases
  487. respond differently to the facial mask.
  488. And so, our digital campaigns
  489. focus on the idea of getting
    the rough consensus through.
  490. So basically, it's a reflection
    of the society,
  491. through Polis, through Slido,
    through the joint platform,
  492. the various tools
    that vTaiwan has prototyped,
  493. we know that people are feeling
    a rough consensus about things
  494. and we're responding
    to the society, saying,
  495. "This is what you all feel
  496. and this is what we're doing
    to respond to your feelings.
  497. And the scientific consensus
    is still developing,

  498. but we know, for example,
  499. people feel that wearing a mask
    mostly protects you,
  500. because it reminds you
    to not touch your face
  501. and wash your hands properly."
  502. And these, regardless of everything else,
  503. are the two things
    that everybody agrees with.
  504. So we just capitalize on that and say,
  505. "OK, wash your hands properly,
  506. and don't touch your face,
  507. and wearing a mask reminds you of that."
  508. And that lets us cut through
  509. the kind of, very ideologically
    charged debates
  510. and focus on what people
    generally resonate with one another.
  511. And that's how we act quickly
    without sacrificing scientific accuracy.
  512. WPR: And this next question
    sort of feels connected to this as well.

  513. It's a question from an anonymous
    community member.
  514. "Pragmatically, do you think
    any of your policies
  515. could be applied in the United States
    under the current Trump administration?"
  516. AT: Quite a few, actually.

  517. We work with many states
    in the US and abroad
  518. on what we call "epicenter
    to epicenter diplomacy." (Laughs)
  519. So what we're doing essentially is,
  520. for example,
    there was a chat bot in Taiwan
  521. that lets you, but especially
    people under home quarantine,
  522. to ask the chat bot anything.
  523. And if there is a scientific adviser
  524. who already wrote
    a frequently asked question,
  525. the chat bot just responds with that,
  526. but otherwise, they will call
    the science advisory board
  527. and write an accessible response to that,
  528. and the spokesdog would translate that
    into a cute dog meme.
  529. And so this feedback cycle

  530. of people very easily accessing,
    finding, and asking a scientist,
  531. and an open API
    that allows for voice assistance
  532. and other third-party developers
    to get through it,
  533. resonates with many US states,
  534. and I think many of them
    are implementing it.
  535. And before the World Health Assembly,
    I think three days before,
  536. we held a 14 countries
    [unclear] lateral meeting,
  537. kind of, pre-WHA,
  538. where we shared many small,
    like, quick wins like this.
  539. And I think many jurisdictions
    took some of that,
  540. including the humor versus rumor.
  541. Many of them said
  542. that they're going to recruit
    comedians now.
  543. WPR: (Laughs) I love that.

  544. DB: I hope so.

  545. WPR: I hope so too.

  546. And we have one more question,
    which is actually a follow-up,
  547. from Michael Backes,
    who asked a question earlier.
  548. "Does the Ministry plan
    to publish their plans in a white paper?"
  549. Sounds like you're already sharing
    your plans with folks,
  550. but do you have a plan
    to put it out on paper?
  551. AT: Of course.

  552. Yeah, and multiple white papers.
  553. So if you go to taiwancanhelp.us,
  554. that is where most of our strategy is,
  555. and that website is actually
    crowdsourced as well,
  556. and it shows that more
    than five million now, I think,
  557. medical masks donated
    to the humanitarian aid.
  558. It's also crowdsourced.
  559. People who have some masks in their homes,
  560. who did not collect the rationed masks,
  561. they can use an app, say,
  562. "I want to dedicate this
    to international humanitarian aid,"
  563. and half of them choose
    to publish their names,
  564. so you can also see the names
    of people who participated in this.
  565. And there's also
    an "Ask Taiwan Anything" website,
  566. (Laughs)

  567. at fightcovid.edu.tw,

  568. that outlines, in white paper form,
    all the response strategies,
  569. so check those out.
  570. WPR: Great.

  571. Well, I will disappear and be back
    later with some other questions.
  572. DB: A blizzard
    of white papers, if you will.

  573. I'd like to turn the focus
    on you a little bit.
  574. How does a conservative anarchist
    become a digital minister?
  575. AT: Yeah, by occupying
    the parliament, and through that.

  576. (Laughs)

  577. More interestingly,

  578. I would say that I go
    working with the government,
  579. but never for the government.
  580. And I work with the people,
    not for the people.
  581. I'm like this Lagrange point
  582. between the people's
    movements on one side
  583. and the government on the other side.
  584. Sometimes right in the middle,
  585. trying to do some coach
    or translation work.
  586. Sometimes in a kind of triangle point,
  587. trying to supply both sides with tools
    for prosocial communication.
  588. But always with this idea
  589. of getting the shared values
    out of different positions,
  590. out of varied positions.
  591. Because all too often,
  592. democracy is built as a showdown
    between opposing values.
  593. But in the pandemic, in the infodemic,

  594. in climate change,
  595. in many of those structural issues,
  596. the virus or carbon dioxide
    doesn't sit down and negotiate with you.
  597. It's a structural issue
    that requires common values
  598. built out of different positions.
  599. And so that is why my working principle
    is radical transparency.
  600. Every conversation, including this one,
  601. is on the record,
  602. including the internal
    meetings that I hold.
  603. So you can see all the different
    meeting transcripts
  604. in my YouTube channel,
    in the SayIt platform,
  605. where people can see,
    after I became digital minister,
  606. I held 1,300 meetings
    with more than 5,000 speakers,
  607. with more than 260,000 utterances.
  608. And every one of them has a URL
  609. that becomes a social object
    that people can have a conversation on.
  610. And because of that,

  611. for example, when Uber's David Plouffe
    visited me to lobby for Uber,
  612. because of radical transparency,
  613. he is very much aware of that,
  614. and so he made all the arguments
    based on public good,
  615. based on sustainability,
    and things like that,
  616. because he knows that the other sides
    would see his positions
  617. very clearly and transparently.
  618. So that encourages people
    to add on each other's argument,
  619. instead of attacking each other's person,
  620. you know, credits and things like that.
  621. And so I think that, more than anything,
  622. is the main principle of conserving
    the anarchism of the internet,
  623. which is about, you know,
  624. nobody can force anyone
    to hook to the internet,
  625. or to adhere to a new internet protocol.
  626. Everything has to be done
    using rough consensus and running code.
  627. DB: I wish you had more counterparts
    all around the world.

  628. Maybe you wish you had more
    counterparts all around the world.
  629. AT: That's why these ideas
    are worth spreading.

  630. DB: There you go.

  631. So one of the challenges that might arise
    with some of these digital tools
  632. is access.
  633. How do you approach that part of it
  634. for folks maybe who don't have
    the best broadband connection
  635. or the latest mobile phone
    or whatever it might be that's required?
  636. AT: Well, anywhere in Taiwan,

  637. even on the top of Taiwan,
    almost 4,000 meters high,
  638. the Saviah, or the Jade Mountain,
  639. you're guaranteed to have
    10 megabits per second
  640. over 4G or fiber or cable,
  641. with just 16 US dollars
    a month, an unlimited plan.
  642. And actually, on the top
    of the mountain, it's faster,
  643. fewer people use that bandwidth.
  644. And if you don't, it's my fault.
  645. It's personally my fault.
  646. In Taiwan, we have broadband
    as a human right.
  647. And so when we're deploying 5G,
  648. we're looking at places
    where the 4G has the weakest signal,
  649. and we begin with those places
    in our 5G deployment.
  650. And only by deploying broadband
    as a human right
  651. can we say that this is for everybody.
  652. That digital democracy
    actually strengthens democracy.
  653. Otherwise, we would be excluding
    parts of the society.
  654. And this also applies to, for example,

  655. you can go to a local
    digital opportunity center
  656. to rent a tablet that's guaranteed
  657. to be manufactured
    in the past three years,
  658. and things like that,
  659. to enable, also,
    the different digital access
  660. by the digital opportunity centers,
    universities and schools,
  661. and public libraries, very important.
  662. And if people who prefer to talk
    in their town hall,
  663. I personally go to that town hall
    with a 360 recorder,
  664. and livestream that to Taipei
    and to other municipalities,
  665. where the central government's
    public servants can join
  666. in a connected room style,
  667. but listening to the local people
    who set the agenda.
  668. So people still do face-to-face meetings,
  669. we're not doing this
    to replace face-to-face meetings.
  670. We're bringing more stakeholders
  671. from central government
    in the local town halls,
  672. and we're amplifying their voices
  673. by making sure the transcripts,
    the mind maps, and things like that
  674. are spread through
    the internet in real time,
  675. but we don't ever ask the elderly to, say,
  676. "Oh, you have to learn typing,
    otherwise you don't do democracy."
  677. It's not our style.
  678. But that requires broadband.
  679. Because if you don't have broadband,
    but only a very limited bandwidth,
  680. you are forced to use
    text-based communication.
  681. DB: That's right.

  682. Well, with access, of course,
  683. comes access for folks
    who maybe will misuse the platform.
  684. You talked a little bit
    about disinformation
  685. and using humor to beat rumor.
  686. But sometimes, disinformation
    is more weaponized.
  687. How do you combat those kinds
    of attacks, really?
  688. AT: Right, so you mean
    malinformation, then.

  689. So essentially, information designed
    to cause intentional public harm.
  690. And that's no laughing matter.
  691. So for that, we have an idea called
    "notice and public notice."
  692. So this is a Reuters photo,
  693. and I will read the original caption.
  694. The original caption says
  695. "A teenage extradition bill
    protester in Hong Kong
  696. is seen during a march to demand democracy
    and political reform in Hong Kong."
  697. OK, a very neutral title by the Reuters.
  698. But there was a spreading
    of malinformation
  699. back last November,
  700. just leading to our presidential election,
  701. that shows something else entirely.
  702. This is the same photo -- that says
  703. "This 13-year-old thug bought new iPhones,
  704. game consoles and brand-name sports shoes,
  705. and recruiting his brothers
    to murder police
  706. and collect 200,000 dollars."
  707. And this, of course, is a weapon
    designed to sow discord,
  708. and to elicit in Taiwan's voters
    a kind of distaste for Hong Kong.
  709. And because they know
    that this is the main issue.

  710. And had we resorted to takedowns,
  711. that will not work,
  712. because that would only
    evoke more outrage.
  713. So we didn't do a takedown.
  714. Instead, we worked with the fact checkers
  715. and professional journalists
  716. to attribute this original message
    back to the first day that it was posted.
  717. And it came from Zhongyang Zhengfawei.
  718. That is the main political and legal unit
    of the central party,
  719. in the Central Communist Party, in CCP.
  720. And we know that it's their Weibo account
    that first did this new caption.
  721. So we sent out a public notice
  722. and with the partners
    in social media companies,
  723. pretty much all of them,
  724. they just put this very small reminder
  725. next to each time that this is shared
    with the wrong caption,
  726. that says "This actually came
    from the central propaganda unit
  727. of the CCP.
  728. Click here to learn more.
    To learn about the whole story."
  729. And that, we found, that has worked,

  730. because people understand
    this is then not a news material.
  731. This is rather an appropriation
    of Reuters' news material
  732. and a copyright infringement
  733. and I think that's part of the [unclear].
  734. In any case, the point
    is that when people understand
  735. that this is an intentional narrative,
  736. they won't just randomly share it.
  737. They may share it,
    but with a comment that says
  738. "This is what the Zhongyang Zhengfawei
    is trying to do to our democracy."
  739. DB: Seems like some
    of the global social media companies

  740. could learn something
    from notice and public notice.
  741. AT: Public notice, that's right.

  742. DB: What advice would you have

  743. for the Twitters and Facebooks
    and LINEs and WhatsApps,
  744. and you name it, of the world?
  745. AT: Yeah.

  746. So, just before our election,
  747. we said to all of them
  748. that we're not making a law
    to kind of punish them.
  749. However, we're sharing
    this very simple fact
  750. that there is this norm in Taiwan
  751. that we even have a separate branch
    of the government,
  752. the control branch,
  753. that published the campaign
    donation and expense.
  754. And it just so occurred to us
  755. that in the previous election,
    the mayoral one,
  756. there was a lot of candidates
  757. that did not include
    any social media advertisements
  758. in their expense to the Control Yuan.
  759. And so essentially, that means
    that there is a separate amount
  760. of political donation and expense
    that evades public scrutiny.
  761. And our Control Yuan
    published their numbers
  762. in raw data form,
  763. that is to say,
    they're not statistics,
  764. but individual records
    of who donated for what cause,
  765. when, where,
  766. and investigative journalists
    are very happy,
  767. because they can then make
    investigative reports
  768. about the connections
    between the candidates
  769. and the people who fund them.
  770. But they cannot work
    with the same material

  771. from the global social media companies.
  772. So I said, "Look, this is very simple.
  773. This is the social norm here,
  774. I don't really care
    about other jurisdictions.
  775. You either adhere to the social norm
    that is set by the Control Yuan
  776. and the investigative journalists,
  777. or maybe you will face social sanctions.
  778. And this is not the government mandate,
  779. but it's the people fed up with,
    you know, black box,
  780. and that's part of the Sunflower
    Occupy's demands, also.
  781. And so Facebook actually published
    in the Ad Library,
  782. I think at that time,
    one of the fastest response strategies,
  783. where everybody who has
    basically any dark pattern advertisement
  784. will get revealed very quickly,
  785. and investigative journalists
    work with the local civic technologists
  786. to make sure that if anybody dare to use
    social media in such a divisive way,
  787. within an hour, there will be
    a report out condemning that.
  788. So nobody tried that during
    the previous presidential election season.
  789. DB: So change is possible.

  790. AT: Mhm.

  791. WPR: Hey there, we have
    some more questions from the community.

  792. There is an anonymous one
  793. that says, "I believe Taiwan
    is outside WHO entirely
  794. and has a 130-part preparation program --
  795. developed entirely on its own --
  796. to what extent does it credit
    its preparation
  797. to building its own system?"
  798. AT: Well, a little bit, I guess.

  799. We tried to warn the WHO,
  800. but at that point --
  801. we are not totally outside,
    we have limited scientific access.
  802. But we do not have any ministerial access.
  803. And this is very different, right?
  804. If you only have limited
    scientific access,
  805. unless the other side's top epidemiologist
    happens to be the vice president,
  806. like in Taiwan's case,
  807. they don't always do
    the storytelling well enough
  808. to translate that into political action
    as our vice president did, right?
  809. So the lack of ministerial
    access, I think,
  810. is to the detriment
    of the global community,
  811. because otherwise,
    people could have responded as we did
  812. in the first day of January,
  813. instead of having to wait for weeks
  814. before the WHO declared
    that this is something,
  815. that there's definitely
    human to human transmission,
  816. that you should inspect people
    coming in from Wuhan,
  817. which they eventually did,
  818. but that's already two weeks
    or three weeks after what we did.
  819. WPR: Makes a lot of sense.

  820. DB: More scientists
    and technologists in politics.

  821. That sounds like that's the answer.
  822. AT: Yeah.

  823. WPR: And then we have another
    question here from Kamal Srinivasan

  824. about your reopening strategy.
  825. "How are you enabling restaurants
    and retailers to open safely in Taiwan?"
  826. AT: Oh, they never closed, so ... (Laughs)

  827. WPR: Oh!

  828. AT: Yeah, they never closed,

  829. there was no lockdown,
    there was no closure.
  830. We just said a very simple thing
    in the CECC press conference,
  831. that there's going to be
    physical distancing.
  832. You maintain one and a half meters indoors
  833. or wear a mask.
  834. And that's it.
  835. And so there are some restaurants
    that put up, I guess, red curtains,
  836. some put very cute teddy bears
    and so on, on the chairs,
  837. to make sure that people spread evenly,
  838. some installed see-through
    glass or plastic walls
  839. between the seats.
  840. There's various social
    innovations happening around.
  841. And I think the only shops
    that got closed for a while,
  842. because they could not innovate
    quick enough to respond to these rules,
  843. was the intimate escort bars.
  844. But eventually,
    even they invented new ways,
  845. by handing out these caps
    that are plastic shielding,
  846. but still leaves room
    for drinking behind it.
  847. And so they opened
    with that social innovation.
  848. DB: That's amazing.

  849. WPR: It is, yeah, it's a lot to learn
    from your strategies there.

  850. Thank you, I'll be back towards the end
    with some final questions.
  851. DB: I'm very happy to hear
    that the restaurants were not closed down,

  852. because I think Taipei
    has some of the best food in the world
  853. of any city that I've visited,
  854. so, you know, kudos to you for that.
  855. So the big concern when it comes
    to using digital tools for COVID
  856. or using digital tools for democracy
  857. is always privacy.
  858. You've talked about that a little bit,
  859. but I'm sure the citizens of Taiwan
  860. are perhaps equally concerned
    about their privacy,
  861. especially given the geopolitical context.
  862. AT: Definitely.

  863. DB: So how do you cope with those demands?

  864. AT: Yeah, we design
    with not only defensive strategy,

  865. like minimization of data collection,
  866. but also proactive measures,
  867. such as privacy-enhancing technologies.
  868. One of the top teams
    that emerged out of our cohack,
  869. the TW response from the Polis,
  870. how to make contact tracing easier,
  871. focused not on the contact tracers,
  872. not on the medical officers,
    but on the person.
  873. So they basically said,
    "OK, you have a phone,
  874. you can record your temperatures,
  875. you can record your whereabouts
    and things like that,
  876. but that is strictly in your phone.
  877. It doesn't even use Bluetooth.
  878. So there's no transmission.
  879. Technology uses open-source,
  880. you can check it,
    you can use it in airplane mode.
  881. And when the contact tracer
    eventually tells you
  882. that you are part of a high-risk group,
  883. and they really want your contact history,
  884. this tool can then generate
    a single-use URL
  885. that only contains
    the precise information,
  886. anonymized,
  887. that the contact tracers want.
  888. But it will not,
    like in a traditional interview,

  889. let you ask --
  890. they ask a question, they only want
    to know your whereabouts,
  891. but you answer with such accuracy
  892. that you end up compromising
    other people's privacy.
  893. So basically, this is about designing
  894. with an aim to enhance
    other people's privacy,
  895. because personal data
    is never truly personal.
  896. It's always social,
    it's always intersectional.
  897. If I take a selfie at a party,
  898. I inadvertently also take
    pretty much everybody else's
  899. who are in the picture, the surroundings,
    the ambiance, and so on,
  900. and if I upload it to a cloud service,
  901. then I actually decimate
    the bargaining power,
  902. the negotiation power
    of everybody around me,
  903. because then their data
    is part of the cloud,
  904. and the cloud doesn't have to
    compensate them
  905. or get their agreement for it.
  906. And so only by designing the tools

  907. with privacy enhancing
    as a positive value,
  908. and not enhancing only
    the person's own privacy,
  909. just like a medical mask, it protects you,
  910. but mostly it also protects others, right?
  911. So if we design tools using that idea,
  912. and always open-source
    and with an open API,
  913. then we're in a much better shape
  914. than in centralized or so-called
    cloud-based services.
  915. DB: Well, you're clearly
    living in the future,

  916. and I guess that's quite literal,
  917. in the sense of,
    it's tomorrow morning there.
  918. AT: Twelve hours.

  919. DB: Yes.

  920. Tell me, what do you see in the future?
  921. What comes next?
  922. AT: Yes, so I see the coronavirus
    as a great amplifier.

  923. If you start with
    an authoritarian society,
  924. the coronavirus,
    with all its lockdowns and so on,
  925. has the potential of making it
    even a more totalitarian society.
  926. If people place their trust, however,
  927. on the social sector,
  928. on the ingenuity of social innovators,
  929. then the pandemic, as in Taiwan,
  930. actually strengthens our democracy,
  931. so that people feel, truly,
    that everybody can think of something
  932. that improves the welfare
    of not just Taiwan,
  933. but pretty much everybody
    else in the world.
  934. And so, my point here

  935. is that the great amplifier
    comes if no matter you want it or not,
  936. but the society, what they can do,
    is do what Taiwan did after SARS.
  937. In 2003, when SARS came,
  938. we had to shut down an entire hospital,
  939. barricading it with no definite
    termination date.
  940. It was very traumatic,
  941. everybody above the age of 30
    remembers how traumatic it was.
  942. The municipalities
  943. and the central government
    were saying very different things,
  944. and that is why after SARS,
  945. the constitutional courts
    charged the legislature
  946. to set up the system as you see today,
  947. and also that is why,
  948. when people responding
    to that crisis back in 2003
  949. built this very robust response system
    that there's early drills.
  950. So just as the Sunflower Occupy,

  951. because of the crisis in trust
    let us build new tools
  952. that put trust first,
  953. I think the coronavirus is the chance
    for everybody who have survived
  954. through the first wave
  955. to settle on a new set of norms
    that will reinforce your founding values,
  956. instead of taking on alien values
    in the name of survival.
  957. DB: Yeah, let's hope so,

  958. and let's hope the rest of the world
    is as prepared as Taiwan
  959. the next time around.
  960. When it comes to digital
    democracy, though,
  961. and digital citizenship,
  962. where do you see that going,
  963. both in Taiwan and maybe
    in the rest of the world?
  964. AT: Well, I have my job description here,

  965. which I will read to you.
  966. It's literally my job description
    and the answer to that question.
  967. And so, here goes.
  968. When we see the internet of things,
  969. let's make it the internet of beings.
  970. When we see virtual reality,
  971. let's make it a shared reality.
  972. When we see machine learning,
  973. let's make it collaborative learning.
  974. When we see user experience,
  975. let's make it about human experience.
  976. And whenever we hear
    the singularity is near,
  977. let us always remember
  978. the plurality is here.
  979. Thank you for listening.
  980. DB: Wow.

  981. I have to give that a little clap,
  982. that was beautiful.
  983. (Laughs)

  984. Quite a job description too.

  985. So, conservative anarchist,
  986. digital minister,
    and with that job description --
  987. that's pretty impressive.
  988. AT: A poetician, yes.

  989. DB: (Laughs)

  990. So I struggle to imagine
  991. an adoption of these techniques in the US,
  992. and that may be my pessimism weighing in.
  993. But what words of hope do you have
    for the US, as we cope with COVID?
  994. AT: Well, as I mentioned,
    during SARS in Taiwan,

  995. nobody imagined we could have
    CECC and a cute spokesdog.
  996. Before the Sunflower movement,
    during a large protest,
  997. there was, I think, half a million
    people on the street, and many more.
  998. Nobody thought that we could have
    a collective intelligence system
  999. that puts open government data
  1000. as a way to rebuild citizen participation.
  1001. And so, never lose hope.
  1002. As my favorite singer, Leonard Cohen --
    a poet, also -- is fond of saying,
  1003. "Ring the bells that still can ring
  1004. and forget any perfect offering.
  1005. There is a crack in everything
    and that is how the light gets in."
  1006. WPR: Wow.

  1007. That's so beautiful,
  1008. and it feels like such a great message
    to, sort of, leave the audience with,
  1009. and sharing the sentiment
  1010. that everyone seems to be so grateful
    for what you've shared, Audrey,
  1011. and all the great information
    and insight into Taiwan's strategies.
  1012. AT: Thank you.

  1013. WPR: And David --

  1014. DB: I was just going to say,
    thank you so much for that,

  1015. thank you for that beautiful
    job description,
  1016. and for all the wisdom you shared
    in rapid-fire fashion.
  1017. I think it wasn't just one idea
    that you shared,
  1018. but maybe, I don't know, 20, 30, 40?
  1019. I lost count at some point.
  1020. AT: Well, it's called
    Ideas Worth Spreading,

  1021. it's a plural form.
  1022. (Laughter)

  1023. DB: Very true.

  1024. Well, thank you so much for joining us.
  1025. WPR: Thank you, Audrey.

  1026. DB: And I wish you luck with everything.

  1027. AT: Thank you, and have a good local time.

  1028. Stay safe.