Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims

Miles Brundage; Shahar  Avin; Jasmine  Wang; Haydn Belfield; Gretchen Krueger; Gillian  Hadfield; Heidy  Khlaaf; Jingying  Yang; Helen  Toner; Ruth Fong; Tegan  Maharaj; Pang Wei  Koh; Sara  Hooker; Jade  Leung; Andrew  Trask; Emma  Bluemke; Jonathan  Lebensold; Cullen  O’Keefe; Mark  Koren; Théo  Ryffel; JB  Rubinovitz; Tamay  Besiroglu; Federica  Carugati; Jack Clark; Peter  Eckersley; Sarah  de Haas; Maritza  Johnson; Ben  Laurie; Alex  Ingerman; Igor  Krawczuk; Amanda  Askell; Rosario  Cammarota; Andrew  Lohn; David  Krueger; Charlotte  Stix; Peter  Henderson; Logan  Graham; Carina  Prunkl; Bianca  Martin; Elizabeth  Seger; Noa  Zilberman; Seán  Ó hÉigeartaigh; Frens Kroeger; Girish  Sastry; Rebecca  Kagan; Adrian  Weller; Brian  Tse; Elizabeth  Barnes; Allan  Dafoe; Paul  Scharre; Ariel  Herbert-Voss; Martijn  Rasser; Shagun  Sodhani; Carrick  Flynn; Thomas Krendl  Gilbert; Lisa  Dyer; Saif  Khan; Yoshua  Bengio; Markus  Anderljung

Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims

Miles Brundage, Shahar Avin, Jasmine Wang, Haydn Belfield, Gretchen Krueger, Gillian Hadfield, Heidy Khlaaf, Jingying Yang, Helen Toner, Ruth Fong, Tegan Maharaj, Pang Wei Koh, Sara Hooker, Jade Leung, Andrew Trask, Emma Bluemke, Jonathan Lebensold, Cullen O’Keefe, Mark Koren, Théo RyffelJB Rubinovitz, Tamay Besiroglu, Federica Carugati, Jack Clark, Peter Eckersley, Sarah de Haas, Maritza Johnson, Ben Laurie, Alex Ingerman, Igor Krawczuk, Amanda Askell, Rosario Cammarota, Andrew Lohn, David Krueger, Charlotte Stix, Peter Henderson, Logan Graham, Carina Prunkl, Bianca Martin, Elizabeth Seger, Noa Zilberman, Seán Ó hÉigeartaigh, Frens Kroeger, Girish Sastry, Rebecca Kagan, Adrian Weller, Brian Tse, Elizabeth Barnes, Allan Dafoe, Paul Scharre, Ariel Herbert-Voss, Martijn Rasser, Shagun Sodhani, Carrick Flynn, Thomas Krendl Gilbert, Lisa Dyer, Saif Khan, Yoshua Bengio, Markus Anderljung

Centre for Trust, Peace and Social Relations

Research output: Contribution to journal › Article › peer-review

Abstract

With the recent wave of progress in artificial intelligence (AI) has come a growing awareness of the large-scale impacts of AI systems, and recognition that existing regulations and norms in industry and academia are insufficient to ensure responsible AI development. In order for AI developers to earn trust from system users, customers, civil society, governments, and other stakeholders that they are building AI responsibly, they will need to make verifiable claims to which they can be held accountable. Those outside of a given organization also need effective means of scrutinizing such claims. This report suggests various steps that different stakeholders can take to improve the verifiability of claims made about AI systems and their associated development processes, with a focus on providing evidence about the safety, security, fairness, and privacy protection of AI systems. We analyze ten mechanisms for this purpose--spanning institutions, software, and hardware--and make recommendations aimed at implementing, exploring, or improving those mechanisms.

Original language	English
Number of pages	80
Journal	ArxiV
Issue number	2004.07213
Publication status	Published - 15 Apr 2020

ASJC Scopus subject areas

Human-Computer Interaction

Access to Document

https://arxiv.org/pdf/2004.07213.pdf

Fingerprint Dive into the research topics of 'Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims'. Together they form a unique fingerprint.

Cite this

Brundage, M., Avin, S., Wang, J., Belfield, H., Krueger, G., Hadfield, G., Khlaaf, H., Yang, J., Toner, H., Fong, R., Maharaj, T., Koh, P. W., Hooker, S., Leung, J., Trask, A., Bluemke, E., Lebensold, J., O’Keefe, C., Koren, M., ... Anderljung, M. (2020). Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims. ArxiV, (2004.07213). https://arxiv.org/pdf/2004.07213.pdf

Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims. / Brundage, Miles; Avin, Shahar ; Wang, Jasmine ; Belfield, Haydn; Krueger, Gretchen; Hadfield, Gillian ; Khlaaf, Heidy; Yang, Jingying ; Toner, Helen ; Fong, Ruth; Maharaj, Tegan ; Koh, Pang Wei ; Hooker, Sara ; Leung, Jade ; Trask, Andrew ; Bluemke, Emma ; Lebensold, Jonathan ; O’Keefe, Cullen ; Koren, Mark; Ryffel, Théo ; Rubinovitz, JB ; Besiroglu, Tamay ; Carugati, Federica ; Clark, Jack; Eckersley, Peter; de Haas, Sarah ; Johnson, Maritza; Laurie, Ben ; Ingerman, Alex ; Krawczuk, Igor ; Askell, Amanda ; Cammarota, Rosario ; Lohn, Andrew ; Krueger, David ; Stix, Charlotte ; Henderson, Peter ; Graham, Logan ; Prunkl, Carina ; Martin, Bianca ; Seger, Elizabeth ; Zilberman, Noa ; Ó hÉigeartaigh, Seán ; Kroeger, Frens; Sastry, Girish ; Kagan, Rebecca ; Weller, Adrian ; Tse, Brian ; Barnes, Elizabeth; Dafoe, Allan ; Scharre, Paul ; Herbert-Voss, Ariel ; Rasser, Martijn ; Sodhani, Shagun ; Flynn, Carrick ; Gilbert, Thomas Krendl ; Dyer, Lisa ; Khan, Saif; Bengio, Yoshua ; Anderljung, Markus .

In: ArxiV, No. 2004.07213, 15.04.2020.

Research output: Contribution to journal › Article › peer-review

Brundage, M, Avin, S, Wang, J, Belfield, H, Krueger, G, Hadfield, G, Khlaaf, H, Yang, J, Toner, H, Fong, R, Maharaj, T, Koh, PW, Hooker, S, Leung, J, Trask, A, Bluemke, E, Lebensold, J, O’Keefe, C, Koren, M, Ryffel, T, Rubinovitz, JB, Besiroglu, T, Carugati, F, Clark, J, Eckersley, P, de Haas, S, Johnson, M, Laurie, B, Ingerman, A, Krawczuk, I, Askell, A, Cammarota, R, Lohn, A, Krueger, D, Stix, C, Henderson, P, Graham, L, Prunkl, C, Martin, B, Seger, E, Zilberman, N, Ó hÉigeartaigh, S, Kroeger, F, Sastry, G, Kagan, R, Weller, A, Tse, B, Barnes, E, Dafoe, A, Scharre, P, Herbert-Voss, A, Rasser, M, Sodhani, S, Flynn, C, Gilbert, TK, Dyer, L, Khan, S, Bengio, Y & Anderljung, M 2020, 'Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims', ArxiV, no. 2004.07213. <https://arxiv.org/pdf/2004.07213.pdf>

Brundage, Miles ; Avin, Shahar ; Wang, Jasmine ; Belfield, Haydn ; Krueger, Gretchen ; Hadfield, Gillian ; Khlaaf, Heidy ; Yang, Jingying ; Toner, Helen ; Fong, Ruth ; Maharaj, Tegan ; Koh, Pang Wei ; Hooker, Sara ; Leung, Jade ; Trask, Andrew ; Bluemke, Emma ; Lebensold, Jonathan ; O’Keefe, Cullen ; Koren, Mark ; Ryffel, Théo ; Rubinovitz, JB ; Besiroglu, Tamay ; Carugati, Federica ; Clark, Jack ; Eckersley, Peter ; de Haas, Sarah ; Johnson, Maritza ; Laurie, Ben ; Ingerman, Alex ; Krawczuk, Igor ; Askell, Amanda ; Cammarota, Rosario ; Lohn, Andrew ; Krueger, David ; Stix, Charlotte ; Henderson, Peter ; Graham, Logan ; Prunkl, Carina ; Martin, Bianca ; Seger, Elizabeth ; Zilberman, Noa ; Ó hÉigeartaigh, Seán ; Kroeger, Frens ; Sastry, Girish ; Kagan, Rebecca ; Weller, Adrian ; Tse, Brian ; Barnes, Elizabeth ; Dafoe, Allan ; Scharre, Paul ; Herbert-Voss, Ariel ; Rasser, Martijn ; Sodhani, Shagun ; Flynn, Carrick ; Gilbert, Thomas Krendl ; Dyer, Lisa ; Khan, Saif ; Bengio, Yoshua ; Anderljung, Markus . / Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims. In: ArxiV. 2020 ; No. 2004.07213.

@article{58e379b32c47458db146be731a8f9e5b,

title = "Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims",

abstract = "With the recent wave of progress in artificial intelligence (AI) has come a growing awareness of the large-scale impacts of AI systems, and recognition that existing regulations and norms in industry and academia are insufficient to ensure responsible AI development. In order for AI developers to earn trust from system users, customers, civil society, governments, and other stakeholders that they are building AI responsibly, they will need to make verifiable claims to which they can be held accountable. Those outside of a given organization also need effective means of scrutinizing such claims. This report suggests various steps that different stakeholders can take to improve the verifiability of claims made about AI systems and their associated development processes, with a focus on providing evidence about the safety, security, fairness, and privacy protection of AI systems. We analyze ten mechanisms for this purpose--spanning institutions, software, and hardware--and make recommendations aimed at implementing, exploring, or improving those mechanisms.",

author = "Miles Brundage and Shahar Avin and Jasmine Wang and Haydn Belfield and Gretchen Krueger and Gillian Hadfield and Heidy Khlaaf and Jingying Yang and Helen Toner and Ruth Fong and Tegan Maharaj and Koh, {Pang Wei} and Sara Hooker and Jade Leung and Andrew Trask and Emma Bluemke and Jonathan Lebensold and Cullen O{\textquoteright}Keefe and Mark Koren and Th{\'e}o Ryffel and JB Rubinovitz and Tamay Besiroglu and Federica Carugati and Jack Clark and Peter Eckersley and {de Haas}, Sarah and Maritza Johnson and Ben Laurie and Alex Ingerman and Igor Krawczuk and Amanda Askell and Rosario Cammarota and Andrew Lohn and David Krueger and Charlotte Stix and Peter Henderson and Logan Graham and Carina Prunkl and Bianca Martin and Elizabeth Seger and Noa Zilberman and {{\'O} h{\'E}igeartaigh}, Se{\'a}n and Frens Kroeger and Girish Sastry and Rebecca Kagan and Adrian Weller and Brian Tse and Elizabeth Barnes and Allan Dafoe and Paul Scharre and Ariel Herbert-Voss and Martijn Rasser and Shagun Sodhani and Carrick Flynn and Gilbert, {Thomas Krendl} and Lisa Dyer and Saif Khan and Yoshua Bengio and Markus Anderljung",

year = "2020",

month = apr,

day = "15",

language = "English",

journal = "ArxiV",

number = "2004.07213",

}

TY - JOUR

T1 - Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims

AU - Brundage, Miles

AU - Avin, Shahar

AU - Wang, Jasmine

AU - Belfield, Haydn

AU - Krueger, Gretchen

AU - Hadfield, Gillian

AU - Khlaaf, Heidy

AU - Yang, Jingying

AU - Toner, Helen

AU - Fong, Ruth

AU - Maharaj, Tegan

AU - Koh, Pang Wei

AU - Hooker, Sara

AU - Leung, Jade

AU - Trask, Andrew

AU - Bluemke, Emma

AU - Lebensold, Jonathan

AU - O’Keefe, Cullen

AU - Koren, Mark

AU - Ryffel, Théo

AU - Rubinovitz, JB

AU - Besiroglu, Tamay

AU - Carugati, Federica

AU - Clark, Jack

AU - Eckersley, Peter

AU - de Haas, Sarah

AU - Johnson, Maritza

AU - Laurie, Ben

AU - Ingerman, Alex

AU - Krawczuk, Igor

AU - Askell, Amanda

AU - Cammarota, Rosario

AU - Lohn, Andrew

AU - Krueger, David

AU - Stix, Charlotte

AU - Henderson, Peter

AU - Graham, Logan

AU - Prunkl, Carina

AU - Martin, Bianca

AU - Seger, Elizabeth

AU - Zilberman, Noa

AU - Ó hÉigeartaigh, Seán

AU - Kroeger, Frens

AU - Sastry, Girish

AU - Kagan, Rebecca

AU - Weller, Adrian

AU - Tse, Brian

AU - Barnes, Elizabeth

AU - Dafoe, Allan

AU - Scharre, Paul

AU - Herbert-Voss, Ariel

AU - Rasser, Martijn

AU - Sodhani, Shagun

AU - Flynn, Carrick

AU - Gilbert, Thomas Krendl

AU - Dyer, Lisa

AU - Khan, Saif

AU - Bengio, Yoshua

AU - Anderljung, Markus

PY - 2020/4/15

Y1 - 2020/4/15

N2 - With the recent wave of progress in artificial intelligence (AI) has come a growing awareness of the large-scale impacts of AI systems, and recognition that existing regulations and norms in industry and academia are insufficient to ensure responsible AI development. In order for AI developers to earn trust from system users, customers, civil society, governments, and other stakeholders that they are building AI responsibly, they will need to make verifiable claims to which they can be held accountable. Those outside of a given organization also need effective means of scrutinizing such claims. This report suggests various steps that different stakeholders can take to improve the verifiability of claims made about AI systems and their associated development processes, with a focus on providing evidence about the safety, security, fairness, and privacy protection of AI systems. We analyze ten mechanisms for this purpose--spanning institutions, software, and hardware--and make recommendations aimed at implementing, exploring, or improving those mechanisms.

AB - With the recent wave of progress in artificial intelligence (AI) has come a growing awareness of the large-scale impacts of AI systems, and recognition that existing regulations and norms in industry and academia are insufficient to ensure responsible AI development. In order for AI developers to earn trust from system users, customers, civil society, governments, and other stakeholders that they are building AI responsibly, they will need to make verifiable claims to which they can be held accountable. Those outside of a given organization also need effective means of scrutinizing such claims. This report suggests various steps that different stakeholders can take to improve the verifiability of claims made about AI systems and their associated development processes, with a focus on providing evidence about the safety, security, fairness, and privacy protection of AI systems. We analyze ten mechanisms for this purpose--spanning institutions, software, and hardware--and make recommendations aimed at implementing, exploring, or improving those mechanisms.

M3 - Article

JO - ArxiV

JF - ArxiV

IS - 2004.07213

ER -