1 of 21

WeTrace

Anonymized contact tracing for COVID-19

Colin Curtin, Laure Merlin, Jose Rivera, Dan Cytermann, Johannes Ebert, Ethan Bialick

❤️

2 of 21

Problem

  • Not enough tests
  • Contact tracing is too slow, manual
  • China relies on control, Singapore on trust

3 of 21

But... why not in my country?

  1. Citizens against surveillance
  2. Citizens don’t trust apps

4 of 21

Solution

5 of 21

Contact tracing,

with care ❤️

6 of 21

  1. Citizen consent

by…

Making it anonymous (the government and others can’t track you)

Making it altruistic (show your care for others)

Making it voluntary

Making it transparent (about how data is used and then discarded)

Solution

7 of 21

2. Citizen trust

by...

Making it transparent (about how data is used and discarded)

Getting recognition from trustworthy institutions (like the WHO, CDC, Universities, high trust companies)

Solution

8 of 21

Our Approach

9 of 21

Our Approach

  • Regularly updated risk score
  • Actionable nudges (testing, hand washing)
  • Send positive test without shame

10 of 21

What’s special?

  • Risk Scoring: With so much unknown and difficult to measure, I want knowledge and action.
  • Anonymity: I want to keep people safe. If I become infected, I don’t want to feel embarrassed. I don’t want to feel like I am being tracked.
  • High Acquisition: I feel safer and more confident in my actions, if I know that everyone is using it.

11 of 21

12 of 21

How it works

13 of 21

  • Risk Scoring Calculation
  • # of geographical contacts in last 14 days
  • Symptoms / testing of those contacts
  • Contact duration

How it works

  • Help Wanted! Need researchers, data scientists to determine risk scores, false positives, false negatives, accuracy.

14 of 21

2. Actionable nudges

  • Advice directly from WHO, CDC
  • Prompts you to get tested

How it works

  • Help Wanted! Need medical professionals to determine highest impact actions, and when

15 of 21

3. Anonymity

  • Collect, anonymize, throw data away
  • Slow down messaging
  • Side benefit: Lowers news anxiety

How it works

  • Help Wanted! Privacy and encryption researchers

Too slow? According to research, a delay of 4 hours is nearly as effective as 0 delay. (Ferretti, Wymant, et al. 2020)

16 of 21

Acquisition / GTM

  • Partnership, testimonial, approval
  • Open source, auditing
  • Channel acquisition: Apple, Google, Facebook
  • Target early adopter, tight knit communities

17 of 21

What can go wrong?

  • Not enough people using it
  • Risk scoring not accurate
  • Negative press (even if untrue!)
  • Unprepared for scale (lots and lots of data)
  • Data breach, deanonymizing locations
  • Low engagement
  • Citizen ignoring advice
  • Ironic effects: will this cause people to not get tested?

18 of 21

Progress So Far, all open source

19 of 21

Next Steps

  • Finish MVP
  • Anonymize data (need cryptography researchers)
  • Acquisition partnerships (companies, NGO’s, media, governments)
  • Collaborate with health researchers

20 of 21

...countries can fight. They can lock down for a few weeks to buy us time, create an educated action plan, and control this virus until we have a vaccine.”

21 of 21

Help Us!

See our Help With Covid page!

See our original Spec Document, with up to date resources

Watch this space for additional ways to join us.