000 07745nam a22005537a 4500
003 NULRC
005 20250519114420.0
008 250516b ph ||||| |||| 00| 0 eng d
022 _a0001-0782
040 _cNULRC
245 _aCommunications of the ACM.
260 _aNew York :
_bAssociation for Computing Machinery (ACM),
_cc2023.
300 _a112 pages :
_bcolor illustrations ;
_c28 cm.
490 _3Communications of the ACM, Volume 66, Issue 1, January 2023
504 _aIncludes bibliographical references.
505 _aComputing Divided: How Wide the Chasm? -- How Not to Win a Tech War -- A Computer Scientist with a Biologist’s Ambition: Advance Humanity -- Making AI Fair, and How to Use It -- Error Control Begins to Shape Quantum Architectures -- The Outlook for Crypto -- Making Traffic a Thing of the Past -- Frederick P. Brooks, Jr. 1931–2022 -- Remembering Valérie Issarny -- From Quantum Computing to Quantum Communications -- Getting a Handle on Handles -- Are Software Updates Useless against Advanced Persistent Threats? -- The End of Programming -- Are We Cobblers without Shoes? Making Computer Science Data FAIR -- The AI Ethicist’s Dirty Hands Problem Distributed Latency Profiling through Critical Path Tracing -- Research for Practice: Crash Consistency -- The Many Faces of Resilience -- ACE: Toward Application-Centric, Edge-Cloud, Collaborative Intelligence -- Democratizing Domain-Specific Computing -- A Linearizability-based Hierarchy for Concurrent Specifications -- The Impact of Auditing for Algorithmic Bias -- Actionable Auditing Revisited— Investigating the Impact of Publicly Naming Biased Performance Results of Commercial AI Products -- Maximal Cocktails
520 _a[Article Title: Computing Divided: How Wide the Chasm?/ Andrew A. Chien, p. 5]
_uhttps://doi.org/10.1145/3572994
520 _a[Article Title: How Not to Win a Tech War/ Moshe Y. Vardi, p. 7]
_uhttps://doi.org/10.1145/357107
520 _a[Article Title: A Computer Scientist with a Biologist's Ambition: Advance Humanity/ Michelle Zhou, p. 9]
_uhttps://doi.org/10.1145/3571064
520 _a[Article Title: Making AI Fair, and How to Use It/ Marc Rotenberg, and Jeremy Roschelle, p. 10-11]
_uhttps://doi.org/10.1145/3570517
520 _a[Article Title: Error Control Begins to Shape Quantum Architectures/ Chris Edwards, p. 13-15]
_uhttps://doi.org/10.1145/3570518
520 _a[Article Title: The Outlook for Crypto/ Neil Savage, p. 16-18]
_b
_uhttps://doi.org/10.1145/3570520
520 _a[Article Title: Making Traffic a Thing of the Past/ Logan Kugler, p. 19-20]
_uhttps://doi.org/10.1145/3570519
520 _a[Article Title: In Memoriam: Frederick P. Brooks, Jr. 1931--2022/ Simson Garfinkel, and Eugene H. Spafford, p. 21-22]
_uhttps://doi.org/10.1145/3572995
520 _a[Article Title: Remembering Valérie Issarny/ John Delaney, p. 23]
_uhttps://doi.org/10.1145/3573217
520 _a[Article Title: From Quantum Computing to Quantum Communications/ Michael A. Cusumano, p. 24-27]
_uhttps://doi.org/10.1145/3571450
520 _a[Article Title: Getting a Handle on Handles/ Alexandra J. Roberts, p. 28-30]
_uhttps://doi.org/10.1145/3571451
520 _a[Article Title: Are Software Updates Useless against Advanced Persistent Threats?/ Fabio Massacci, and Giorgio di Tizio, p. 31-33]
_uhttps://doi.org/10.1145/3571452
520 _a[Article Title: The End of Programming/ Matt Welsh, p. 34-35]
_uhttps://doi.org/10.1145/3570220
520 _a[Article Title: Are We Cobblers without Shoes?: Making Computer Science Data FAIR/ Natasha Noy, and Carole Goble, p. 36-38]
_uhttps://doi.org/10.1145/3528574
520 _a[Article Title: The AI Ethicist's Dirty Hands Problem/ Henrik Skaug Sætra, Mark Coeckelbergh, and John Danaher, p. 39-41]
_uhttps://doi.org/10.1145/35297
520 _a[Article Title: Distributed Latency Profiling through Critical Path Tracing/ Brian Eaton, Jeff Stewart, Jon Tedesco, and N. Cihan Tas, p. 44-51]
_uhttps://doi.org/10.1145/3570522
520 _a[Article Title: Research for Practice: Crash Consistency/Ramnatthan Alagappan, and Peter Alvaro, p. 52-54]
_uhttps://doi.org/10.1145/3570521
520 _a[Article Title: The Many Faces of Resilience/Ted G. Lewis, p. 56-61]
_uhttps://doi.org/10.1145/3519262
520 _a[Article Title: ACE: Toward Application-Centric, Edge-Cloud, Collaborative Intelligence/Luhui Wang, Cong Zhao, Shusen Yang, and Xinyu Yang, p. 62-73]
_uhttps://doi.org/10.1145/3529087
520 _a[Article Title: Democratizing Domain-Specific Computing/ Yuze Chi, Weikang Qiao, Atefeh Sohrabizadeh, Jie Wang, and Jason Cong, p. 74-85]
_b
_uhttps://doi.org/10.1145/3524108
520 _a[Article Title: A Linearizability-based Hierarchy for Concurrent Specifications/ Armando Castañeda, Sergio Rajsbaum, and Michel Raynal, p. 86-97]
_uhttps://doi.org/10.1145/3546826
520 _a[Article Title: Technical Perspective: The Impact of Auditing for Algorithmic Bias/ Vincent Conitzer, Gillian K. Hadfield, and Shannon Vallor, p. 100]
_uhttps://doi.org/10.1145/3571152
520 _a[Article Title: Actionable Auditing Revisited: Investigating the Impact of Publicly Naming Biased Performance Results of Commercial AI Products/ Inioluwa Deborah Raji, and Joy Buolamwini, p. 101-108] Abstract: Although algorithmic auditing has emerged as a key strategy to expose systematic biases embedded in software platforms, we struggle to understand the real-world impact of these audits and continue to find it difficult to translate such independent assessments into meaningful corporate accountability. To analyze the impact of publicly naming and disclosing performance results of biased AI systems, we investigate the commercial impact of Gender Shades, the first algorithmic audit of gender- and skin-type performance disparities in commercial facial analysis models. This paper (1) outlines the audit design and structured disclosure procedure used in the Gender Shades study, (2) presents new performance metrics from targeted companies such as IBM, Microsoft, and Megvii (Face++) on the Pilot Parliaments Benchmark (PPB) as of August 2018, (3) provides performance results on PPB by non-target companies such as Amazon and Kairos, and (4) explores differences in company responses as shared through corporate communications that contextualize differences in performance on PPB. Within 7 months of the original audit, we find that all three targets released new application program interface (API) versions. All targets reduced accuracy disparities between males and females and darker- and lighter-skinned subgroups, with the most significant update occurring for the darker-skinned female subgroup that underwent a 17.7--30.4% reduction in error between audit periods. Minimizing these disparities led to a 5.72--8.3% reduction in overall error on the Pilot Parliaments Benchmark (PPB) for target corporation APIs. The overall performance of non-targets Amazon and Kairos lags significantly behind that of the targets, with error rates of 8.66% and 6.60% overall, and error rates of 31.37% and 22.50% for the darker female subgroup, respectively. This is an expanded version of an earlier publication of these results, revised for a more general audience, and updated to include commentary on further developments.
_uhttps://doi.org/10.1145/3571151
520 _a[Article Title: Maximal Cocktails/ Dennis Shasha, p. 112]
_uhttps://doi.org/10.1145/3571275
690 _aARTIFICIAL INTELLIGENCE
690 _aBIOMETRICS
690 _aCOMPUTING INDUSTRY
690 _aDESIGNING SOFTWARE
690 _aINTERACTIVE COMPUTATION
690 _aINTERACTION PARADIGMS
690 _aMACHINE LEARNING
942 _2lcc
_cSER
_n0
999 _c2218
_d2218