ROGUEWOLF. SmartCities: Anticipating Agents of Change. Adam Amos-Binks Colleen Stacy Lucia Titus Kathleen Vogel Lori Wachter.

Transcription

1 ROGUEWOLF SmartCities: Anticipating Agents of Change Adam Amos-Binks Colleen Stacy Lucia Titus Kathleen Vogel Lori Wachter November 2, 2016

2 Outline Motivation: SmartCities + Anticipatory thinking Approach Background: SmartCities Anticipatory thinking Intentional planning Future oriented SATs Case Study: SmartCities Agents Summary

3 SmartCities IoT - $4.6 Trillion value-at-stake for public sector1 Cities will generate 63% of total value-at-stake City infrastructure transformed through IoT 1 - J Bradley, C Reberger, A Dixit, and V Gupta Internet of Everything: A $4.6 Trillion Public-Sector Opportunity. Cisco, no. 23: 1 17.

4 Anticipatory Thinking (AT) Deliberately thinking about the future2 2 - Klein, G., D. Snowden, and C. L. Pin nticipatory ThinkingA. KL Mosier, & UM Fischer, Informed by Knowledge (2011)

5 Approach Support Anticipatory Thinking (trajectory tracking) with Artificial Intelligence (intentional planning) and structured analytic techniques (AFA + backcasting) to produce Anticipatory Intelligence for an emerging use case (SmartCities)

6 Background

7 Smart Cities Interconnected power grids reducing power waste Smarter transportation resulting in increased traffic management Smarter infrastructures that reduce hazards and increase efficiency But also means: Expanded attack surface for malicious actors Cyber vulnerabilities resulting in physical consequences

8 Smart Cities Case Examples: Ukraine Power Grid Hack: Cyber Used in Warfare (Dec. 2015) OVH WebCam Attack: 1Tbps DDoS (Sept. 2016) Dyn DDoS: IoT->DNS Attack (Oct. 2016) Ideal for AT as it s coming, still time to have an impact! What entities will influence the value at stake?

9 Anticipatory Thinking Different from prediction, relies on expertise Generate a view of the future Identified forms: Pattern Matching Convergent Thinking Trajectory Tracking: Narrative...is of particular use in trajectory tracking to understand the possible moves of other actors in the decision space - Klein et al. (2011) 2 - Klein, G., D. Snowden, and C. L. Pin Anticipatory Thinking. KL Mosier, & UM Fischer, Informed by Knowledge (2011)

10 Anticipatory Thinking Trajectory Tracking Actors actions influence our decision space Does not scale, hard to incorporate new actors

11 Artificial Intelligence Russell and Norvig s AI taxonomy Thinking Humanly Thinking Rationally Acting Humanly Acting Rationally

12 Artificial Intelligence Automated planning: Industrial problems Plans: Theorized to represent plots3 Lead to intentional planning 3 - Schank, R, Abelson, R Scripts, plans, goals and understanding: An inquiry into human knowledge structures.

13 Intentional Planning Intentional path Character actions are believable, goal-oriented Motivated when goals are not the current state Intentional plan Consists of intentional paths Character goals conflict + collaborate 4 - Riedl, M. Young, R. M Narrative Planning: Balancing Plot and Character Journal of Artificial Intelligence Research

14 Intentional Planning Need library of intentional paths!

15 Future-oriented SATs What are SATs? Mechanisms to guide analysts to think about a problem. Why use SATs? Externalize and decompose thinking Help analyst mitigate cognitive limitations Enable peer review

16 Future-oriented SATs Alternative Futures Analysis Identified alternative trajectories Based on critical uncertainties Inform and illuminate decisions, plans and actions today Backcasting Scenario analysis Desirable futures Plot a point in time, work backward to present

17 Case Study Knowledge Engineering

18 Alternative Futures Analysis How might the world of the Internet evolve? Bottom line uncertainty Corporate Bliss (corporate dominance with little risk) Shangri La Security of the Internet Regulations without bite (government regulates to no avail) Government Clockwork Orange (hackers have leverage ) Involvement Independent Nation State Security (increasing regulatory control) Leviathan 5 - Healey, Jason, and Hughes Barry Risk Nexus: Overcome by Cyber Risks? Economic Benefits and Costs of Alternate Cyber Futures. The Atlantic Group.

19 Internet Evolution Axes: Security Shangri La - Clockwork Orange Gov. Involvement Independent Leviathan

20 Industry Sector Actors Sectors: Electricity Water Transportation

21 Strategic Plans Strategic plans Intentional plans Corporate strategies (e.g. Apple, Duke Energy, Toyota) Government future vision policy docs (e.g. Raleigh 2030) NGO/NPO strategies (e.g. WRF) Non-state actor strategies (e.g. ISIS 2020)

22 Intentional Paths Actor: Water Research Foundation (NGO) Sector: Water Goal: Shangri-La (H), Leviathan (H)

23 Intentional Paths Actor: ISIS (Militant) Sector: Transportation Goal: Clockwork Orange (H), Leviathan (H)

24 Intentional Paths Actor: Government Sector: Transportation Goal: Shangri-La (M), Independent (L)

25 Insights (Cooperation) Water Research Foundation & Government, water Leviathan

26 Insights (Conflict) ISIS vs Government, Transportation

27 Use Case Hand crafted knowledge engineering SmartCities actors (sectors from DHS) SATs (AFA, backcasting) Strategic plans Intentional paths Trajectory tracking (conflicts, collaboration)

28 Summary Trajectory tracking w/ intentional planning SmartCity knowledge engineering using SATs Strategic plans intentional paths

29 Next Steps DO7 OpenKE: automate strategic intentional paths Input for trajectory tracking in serious game

30 Discussion Questions? {aaamosbi, cstacy, lltitus, kmvogel,