The IEEE Global Initiative for Ethical Considerations in Artificial Intelligence and Autonomous Systems. Overview April, 2017

Transcription

1 The IEEE Global Initiative for Ethical Considerations in Artificial Intelligence and Autonomous Systems Overview April, 2017

2 @johnchavens

3 3 IEEE Standards Association

4 IEEE s Technology Ethics Landscape Code of Ethics Professional Guidelines Help define intended behavior by professionals in the field IEEE Code of Ethics Software Engineering CoE Recent AI Now and White House Reports Professional Ethics IEEE Ad Hoc Committee on Ethics Programs Technology Ethics Behavioral Impacts Impact of (and response to) professional behaviors in the context of the guidelines Research Ethics and Plagiarism Business Ethics Safety/labor/equality considerations and policy Ethics Education Technology Impacts Ethical/societal impacts of the technologies themselves IEEE Global Initiative for Ethical Considerations in Artificial Intelligence and Autonomous Systems Ad Hoc on Design for Ethics Events/publications 4

5

6 6 Ethically Aligned Design A Vision for Prioritizing Human Wellbeing with Artificial Intelligence and Autonomous Systems

7 Overview Discussions / Ad Hoc Work begun in late 2015 Public launch as an Industry Connections Group, April 2016 Currently over two hundred AI / Ethics experts involved from the US, EU, Australia, India, China, Korea and Japan 13 Committees creating content plus multiple supporting groups Actively working on Ethically Aligned Design, version 2 Seven approved Standards Working Groups so far Events featuring Ethically Aligned Design in India, Japan, China, World Economic Forum, EU Parliament, and UN/ITU in Geneva 7

8 History Why / When the IEEE Global Initiative Was Created: To move beyond both the fear and the uncritical admiration regarding autonomous and intelligent technologies. To show that aligning technology with ethical values will help advance innovation with these new tools while diminishing fear in the process. Discussions / Ad Hoc Work begun in late 2015 Public launch April

9 The Car Trap I am concerned about blind faith in technical progress, a progress "that takes us to a destination that none of us actually want: loss of personal autonomy, loss of skills and employment, loss of knowledge about how the world works. 9

10 Mandate Create Ethically-aligned autonomous and intelligent systems: Incorporate ethical aspects of human wellbeing that may not automatically be considered in the current design and manufacture of AIS technologies. To reframe the notion of success so human progress can include the intentional prioritization of individual, community, and societal ethical values. 10

11 Mission To ensure every technologist is educated, trained, and empowered to prioritize ethical considerations in the design and development of autonomous and intelligent systems. 11

12 IEEE-SA Standards Projects IEEE P7000: Model Process for Addressing Ethical Concerns During System Design IEEE P7001: Transparency of Autonomous Systems IEEE P7002: Data Privacy Process IEEE P7003: Algorithmic Bias Considerations IEEE P7004: Standard on Child and Student Data Governance IEEE P7005: Standard on Employer Data Governance IEEE P7006: Standard on Personal Data AI Agent Working Group 12

13 IEEE-SA Standards Projects P7000 Model Process for Addressing Ethical Concerns During System Design outlines an approach for identifying and analyzing potential ethical issues in a system or software program from the onset of the effort. The values-based system design methods addresses ethical considerations at each stage of development to help avoid negative unintended consequences while increasing innovation. 13

14 IEEE-SA Standards Projects P7001 Transparency of Autonomous Systems provides a Standard for developing autonomous technologies that can assess their own actions and help users understand why a technology makes certain decisions in different situations. For example, if a selfdriving car chooses to brake unexpectedly it should inform the driver what the reason was for braking because something blocked its path. Or if a caregiver robot inadvertently gives a patient the wrong medication, the robot must explain how it made that mistake so the error can be corrected. The project also offers ways to provide transparency and accountability for a system to help guide and improve it, such as incorporating an event data recorder in a self-driving car or accessing data from a device s sensors. 14

15 IEEE-SA Standards Projects P7002 Data Privacy Process specifies how to manage privacy issues for systems or software that collect personal data. It will do so by defining requirements that cover corporate data collection policies policies and quality assurance. It also includes a use case and data model for organizations developing applications involving personal information. The standard will help designers by providing ways to identify and measure privacy controls in their systems utilizing privacy impact assessments. 15

16 IEEE-SA Standards Projects P7003 Algorithmic Bias Considerations provides developers of algorithms for autonomous or intelligent systems with protocols to avoid negative bias in their code. Bias could include the use of subjective or incorrect interpretations of data like mistaking correlation with causation. The project offers specific steps to take for eliminating issues of negative bias in the creation of algorithms. The standard will also include benchmarking procedures and criteria for selecting validation data sets, establishing and communicating the application boundaries for which the algorithm has been designed, and guarding against unintended consequences. 16

17 IEEE-SA Standards Projects P7004 Standard for Child and Student Data Governance provides processes and certifications for transparency and accountability for educational institutions that handle data meant to ensure the safety of students. The standard defines how to access, collect, share, and remove data related to children and students in any educational or institutional setting where their information will be access, stored, or shared. 17

18 IEEE-SA Standards Projects P7005 Standard for Transparent Employer Data Governance provides guidelines and certifications on storing, protecting, and using employee data in an ethical and transparent way. The project recommends tools and services that help employees make informed decisions when their personal information. The standard will help provide clarity and recommendations both for how employees can share their information in a safe and trusted environment as well as how employers can align with employees in this process while still utilizing information needed for regular work flows. The standard is modeled after the European Union s General Data Protection Regulation, which outlines how to secure an individual s data within the EU. 18

19 IEEE-SA Standards Projects P7006 Standard for Personal Data Artificial Intelligence (AI) Agent addresses concerns raised about machines making decisions without human input. This standard hopes to educate government and industry on why it s best to put mechanisms into place to enable the design of systems that will mitigate the ethical concerns when AI systems can organize and share personal information on their own. Designed as a tool to allow any individual to essentially create their own personal terms and conditions for their data, the AI Agent will provide a technological tool for individuals to manage and control their identity in the digital and virtual world. 19

20 20 Mumbai

21 21 China

22 22 Japan

23 23 Work Process

24 24 Meetings

25 Organization - Committees Executive Committee Global Oversight and Strategic Leadership Content Committees Create content for EAD / Standards Projects Includes Thirteen Content Committees Advisory Committees Ongoing advisory role to Content Committees Overseen by Executive Committee 25

26 Thank you! The Global Initiative for Ethical Considerations in Artificial Intelligence and Autonomous Systems Contact: John C. Havens, Executive Director /