PriBots Conversational Privacy with Chatbots Hamza Harkous 1 , - - PowerPoint PPT Presentation

▶

pribots

PriBots Conversational Privacy with Chatbots Hamza Harkous 1 , - - PowerPoint PPT Presentation

Aug 20, 2022 201 likes •711 views

PriBots Conversational Privacy with Chatbots Hamza Harkous 1 , Kassem Fawaz 2 , Kang. G. Shin 2 , Karl Aberer 1 1 EPFL; 2 University of Michigan Workshop on the Future of Privacy Notices and Indicators, SOUPS 2016 Privacy Notice: Current State 2

slide-1

SLIDE 1

PriBots

Hamza Harkous1, Kassem Fawaz2, Kang. G. Shin2, Karl Aberer1

1EPFL; 2University of Michigan

Conversational Privacy with Chatbots

Workshop on the Future of Privacy Notices and Indicators, SOUPS 2016

slide-2

SLIDE 2

2

Privacy Notice: Current State

slide-3

SLIDE 3

legally binding human understandable

Dual Role:

2

Privacy Notice: Current State

slide-4

SLIDE 4

Can we split these roles?

legally binding human understandable

Dual Role:

2

Privacy Notice: Current State

slide-5

SLIDE 5

Standardization

3

slide-6

SLIDE 6

Standardization

3

Summarization

slide-7

SLIDE 7

Standardization

3

Summarization

Challenges

ne size fits all

user education

slide-8

SLIDE 8

4

Privacy Choice: Current State

slide-9

SLIDE 9

fragmented ecosystem difgicult to find

4

Privacy Choice: Current State

slide-10

SLIDE 10

5

Conversation-first Interfaces

slide-11

SLIDE 11

The Rise of Conversational UI

6

slide-12

SLIDE 12

PriBots: Conversational Privacy Bots

Message | 7

slide-13

SLIDE 13

PriBots: Conversational Privacy Bots

Message | 7

slide-14

SLIDE 14

PriBots: Conversational Privacy Bots

Message |

Appeal to new tech adopters

7

slide-15

SLIDE 15

PriBots: Conversational Privacy Bots

Message |

Appeal to new tech adopters Appeal to existing users

7

slide-16

SLIDE 16

PriBots: Conversational Privacy Bots

Message |

Appeal to new tech adopters Appeal to existing users

An intuitive way to 1. communicate privacy policies 2. adjust privacy preferences

7

slide-17

SLIDE 17

1- Communicating Privacy Policies

8

slide-18

SLIDE 18

Channel

9

slide-19

SLIDE 19

Channel

Primary

9

slide-20

SLIDE 20

Channel

Primary Secondary

9

slide-21

SLIDE 21

Timing

10

slide-22

SLIDE 22

Timing

At-setup

10

slide-23

SLIDE 23

Timing

At-setup On-demand

10

slide-24

SLIDE 24

F e e d b a c k

implicit: sentiment analysis explicit: structured messages gathering users’ concerns

11

slide-25

SLIDE 25

Voicing User Concerns

Providers traditionally say what they want

12

slide-26

SLIDE 26

Voicing User Concerns

Providers traditionally say what they want Users’ concerns might not be covered

12

slide-27

SLIDE 27

Voicing User Concerns

Providers traditionally say what they want Users’ concerns might not be covered PriBots activate the two-way channel

12

slide-28

SLIDE 28

2- Setting Privacy Preferences

13

slide-29

SLIDE 29

14

slide-30

SLIDE 30

Service and platform-dependent interface

14

slide-31

SLIDE 31

Service and platform-dependent interface Tradeofgs for simplicity: try finding this setting on Mobile Web version

14

slide-32

SLIDE 32

15

slide-33

SLIDE 33

Unique interface with all functionalities Ability to suggest adjustments to the user (combining notice and choice/preferences )

15

slide-34

SLIDE 34

User Input Analysis & Classification

System Architecture

16

slide-35

SLIDE 35

User Input Analysis & Classification Query

Structured Query

Statement Yes No

System Architecture

16

slide-36

SLIDE 36

User Input Analysis & Classification Query

Structured Query

Statement Yes No

Retrieval Module

Result

Knowledge Base

System Architecture

16

slide-37

SLIDE 37

User Input Analysis & Classification Query

Structured Query

Statement Yes No

Confident?

Answer Formulation in NL Fallback Answer   Generation

Yes No

Retrieval Module

Result

Knowledge Base

System Architecture

16

slide-38

SLIDE 38

PriBot Reply User Input Analysis & Classification Query

Structured Query

Statement Yes No

Confident?

Answer Formulation in NL Fallback Answer   Generation

Yes No

Retrieval Module

Result

Knowledge Base

System Architecture

16

slide-39

SLIDE 39

PriBot Reply User Input Analysis & Classification Query

Structured Query

Statement Yes No

Confident?

Answer Formulation in NL Fallback Answer   Generation

Yes No

Retrieval Module

Result

Knowledge Base

System Architecture

16

Feedback DB

Analytics Amendments/ Improvements

Augment the Knowledge Base

unanswered queries
frequent questions
user sentiments

Feedback DB

slide-40

SLIDE 40

Challenges

17

slide-41

SLIDE 41

Mature User Understanding

Text processing Question answering Domain-specific datasets and ontologies Graceful fallback

18

slide-42

SLIDE 42

Legal Challenges

Inherently error prone: are they legally binding? Accounting for false-positives and false-negatives The case of 3rd party PriBots: defamation possibilities?

19

slide-43

SLIDE 43

Trusting the Machine

rule-based vs. AI-based user backlash? regulate the confidence level

20

slide-44

SLIDE 44

PriBots’ Personality

positive tone → higher trust diversified content → reduced habituation

21

slide-45

SLIDE 45

D e p l

y

m e n t

22

slide-46

SLIDE 46

provider 3rd parties

D e p l

y

m e n t

22

slide-47

SLIDE 47

provider 3rd parties

D e p l

y

m e n t

22

Suitable for Voice Assistants

slide-48

SLIDE 48

Rule-based Prototype System Implementation User studies

What’s Next?

23

Privacy as a Dialogue

slide-49

SLIDE 49

24

Questions/Feedback?

hamza.harkous@gmail.com hamzaharkous.com

slide-50

SLIDE 50

Image/Media Credits

Zara Picken: slide 12 Egor Kosten: slide 24 Alex Prokhoda: slide 6 Freepik: slide 23 Geofg Keough: slide 14 Victor: slide 12