19. How to implement data privacy? A conversation with Klaudius Kalcher, cofounder and chief data scientist of MOSTLY AI

1

Introduction

00:00 • 3min

2

What's the Story Behind the Company?

02:31 • 3min

3

How Data Sharing Boosted My Researches

05:02 • 3min

4

The Passion for Privacy

08:02 • 3min

5

Data Democracy - A Practical Problem

10:40 • 2min

6

Syntatic Data - What's Your Favorite Use Case?

12:13 • 3min

7

The Privacy of Symthatic Data

15:13 • 2min

8

Privacy Is an Important Thing for All of Us

17:18 • 3min

9

How to Identify a Thing With Only Four Data Points

20:42 • 2min

10

Why Do We Need Synthetic Data in the Legacy a Portfolio?

22:38 • 2min

11

The Privacy Utility Tradef Is Really Bad

24:18 • 4min

12

Is All Synthetic Data Automatically Private by Design?

28:04 • 5min

13

Synthetic Data - What Are the Privacy Threats?

32:35 • 3min

14

How to Predict a Person's Age and Whether She's Male or Female

35:19 • 2min

15

Is There a Privacy Leakage?

36:49 • 4min

16

How to Identify a Six Year Old Single Single Person?

40:25 • 2min

17

Synthetic Data in the City of Vienna

42:05 • 2min

18

Generate Synthetic Data With Domain Agnostic Generating

44:11 • 3min

19

What's Differential Privacy?

47:07 • 2min

20

The Flaw of Caonanimity

48:58 • 2min

21

The Limitations of Differential Privacy in Synthetic Data

50:32 • 4min

22

Is There a Limitation to Differential Privacy?

54:39 • 2min

23

Is There a Level of Privacy Guarantee?

56:13 • 2min

24

Synthetic Data and Differential Privacy

58:35 • 3min

25

Data Democratization Podcast Interview - What Are the Most Important Takeaways?

01:01:11 • 3min

26

How to Protect Your Data With Synthetic Data

01:03:56 • 4min