The use of emotion recognition technologies in the workplace is expanding. These technologies claim to provide insights into internal emotional states based on external cues like facial expressions. Despite interconnections between autism and the development of emotion recognition technologies as reported in prior research, little attention has been paid to the particular issues that arise for autistic individuals when emotion recognition technologies are implemented in consequential settings like the workplace. This article examines recent literature on autism and on emotion recognition technologies to argue that the risks of the use of emotion recognition technologies in the workplace are heightened for autistic people. Following a brief overview of emotion recognition technologies, this argument is made by focusing on the issues that arise through the development and deployment of emotion recognition technologies. Issues related to the development of emotion recognition technologies include fundamental problems with the science behind the technologies, the underrepresentation of autistic individuals in data sets and the problems with increasing this representation, and annotation of the training data for the technologies. Issues related to implementation include the invasive nature of emotion recognition technologies, the sensitivity of the data used, and the imposition of neurotypical norms on autistic workers through their use. The article closes with a call for future research on the implications of these emergent technologies for autistic individuals.
Lay abstract
Technologies using artificial intelligence to recognize people’s emotional states are increasingly being developed under the name of emotional recognition technologies. Emotion recognition technologies claim to identify people’s emotional states based on data, like facial expressions. This is despite research providing counterevidence that emotion recognition technologies are founded on bad science and that it is not possible to correctly identify people’s emotions in this way. The use of emotion recognition technologies is widespread, and they can be harmful when they are used in the workplace, especially for autistic workers. Although previous research has shown that the origins of emotion recognition technologies relied on autistic people, there has been little research on the impact of emotion recognition technologies on autistic people when it is used in the workplace. Through a review of recent academic studies, this article looks at the development and implementation processes of emotion recognition technologies to show how autistic people in particular may be disadvantaged or harmed by the development and use of the technologies. This article closes with a call for more research on autistic people’s perception of the technologies and their impact, with involvement from diverse participants.
AEON Retail. (2024, July1). AEON news release: 世界初!エッジAI を従業員の笑顔·発声トレーニングに導入 従業員の笑顔と挨拶の実施率が1.6 [A global First!—Implementation of Edge AI in smile and voice training for employees leads to a 1.6x increase in smiles and greetings from employees]. https://www.aeonretail.jp/pdf/240701R_1.pdf
3.
AndalibiN.BussJ. (2020). The human in emotion recognition on social media: Attitudes, outcomes, risks. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems (pp. 1–16). Association for Computing Machinery. https://doi.org/10.1145/3313831.3376680
4.
AndersonC.ButtC.SarsonyC. (2021). Young adults on the autism spectrum and early employment-related experiences: Aspirations and obstacles. Journal of Autism and Developmental Disorders, 51(1), 88–105.
AwumeyE.DasS.ForlizziJ. (2024). A systematic review of biometric monitoring in the workplace: Analyzing socio-technical harms in development, deployment and use. In The 2024 ACM Conference on Fairness, Accountability, and Transparency (pp. 920–932). Association for Computing Machinery. https://doi.org/10.1145/3630106.3658945
7.
BakirV.GhotbiN.HoT. M.LafferA.MantelloP.McStayA.MirandaD.MiyashitaH.PodoletzL.TanakaH.UrquhartL. (2022). Emotional AI in cities: Cross-cultural lessons from the UK and Japan on designing for an ethical life. In CartaS. (Ed.), Machine learning and the city: Applications in architecture and urban design (pp. 621–624). John Wiley & Sons Ltd.
8.
BarrettL. F. (2016). Navigating the science of emotion. In MeiselmanH. L. (Eds.), Emotion measurement (pp. 39–84). Elsevier.
9.
BarrettL. F.AdolphsR.MarsellaS.MartinezA. M.PollakS. D. (2019). Emotional expressions reconsidered: Challenges to inferring emotion from human facial movements. Psychological Science, 20, 1–68.
10.
BenjaminR. (2019). Race after technology. Polity Press.
11.
BoydK. L.AndalibiN. (2023). Automated emotion recognition in the workplace: How proposed technologies reveal potential futures of work. Proceedings of the ACM on Human-Computer Interaction, 7(CSCW1), 1–37. https://doi.org/10.1145/3579528
12.
BrewerR.BiottiF.CatmurC.PressC.HappéF.CookR.BirdG. (2016). Can neurotypical individuals read autistic facial expressions? Atypical production of emotional facial expressions in autism spectrum disorders: Expression production in autism. Autism Research, 9(2), 262–271. https://doi.org/10.1002/aur.1508
13.
BuryS. M.FlowerR. L.ZullaR.NicholasD. B.HedleyD. (2021). Workplace social challenges experienced by employees on the autism spectrum: An international exploratory study examining employee and supervisor perspectives. Journal of Autism and Developmental Disorders, 51(5), 1614–1627.
14.
CageE.Troxell-WhitmanZ. (2019). Understanding the Reasons, contexts and costs of camouflaging for autistic adults. Journal of Autism and Developmental Disorders, 49(5), 1899–1911. https://doi.org/10.1007/s10803-018-03878-x
15.
CaiR. Y.RichdaleA. L.UljarevićM.DissanayakeC.SamsonA. C. (2018). Emotion regulation in autism spectrum disorder: Where we are and where we need to go: ER in ASD: Where we are & where we need to go. Autism Research, 11(7), 962–978. https://doi.org/10.1002/aur.1968
16.
CassidyS.BradleyP.RobinsonJ.AllisonC.McHughM.Baron-CohenS. (2014). Suicidal ideation and suicide plans or attempts in adults with Asperger’s syndrome attending a specialist diagnostic clinic: A clinical cohort study. The Lancet Psychiatry, 1(2), 142–147. https://doi.org/10.1016/S2215-0366(14)70248-2
17.
ConnerC. M.GoltJ.ShafferR.RighiG.SiegelM.MazefskyC. A. (2021). Emotion dysregulation is substantially elevated in autism compared to the general population: Impact on psychiatric services. Autism Research, 14(1), 169–181. https://doi.org/10.1002/aur.2450
18.
CrawfordK. (2021a). Atlas of AI: Power, politics, and the planetary costs of artificial intelligence. Yale University Press.
19.
CrawfordK. (2021b). Time to regulate AI that interprets human emotions. Nature, 592, 167.
FerraroA. (2020). Affective rights: A foundation for ethical standards. In 2020 IEEE International Symposium on Technology and Society (ISTAS) (pp. 1–11). IEEE. https://doi.org/10.1109/ISTAS50296.2020.9462172
24.
Garcia-GarciaJ. M.PenichetV. M. R.LozanoM. D.FernandoA. (2022). Using emotion recognition technologies to teach children with autism spectrum disorder how to identify and express emotions. Universal Access in the Information Society, 21(4), 809–825. https://doi.org/10.1007/s10209-021-00818-y
25.
GhotbiN.HoM. T. (2021). Moral awareness of college students regarding artificial intelligence. Asian Bioethics Review, 13(4), 421–433. https://doi.org/10.1007/s41649-021-00182-2
26.
GreeneG. (2020). The ethics of AI and emotional intelligence: Data sources, applications, and questions for evaluating ethics risk. Partnership on AI.
27.
GroteT.KornO. (2017). Risks and potentials of affective computing. Why the ACM Code of Ethics requires a substantial revision. In CHI 2017 workshop on Ethical Encounters in HCI.
28.
HernandezJ.LovejoyJ.McDuffD.SuhJ.O’BrienT.SethumadhavanA.GreeneG.PicardR.CzerwinskiM. (2021, 28September–1October). Guidelines for assessing and minimizing risks of emotion recognition applications [Conference session]. 2021 9th International Conference on Affective Computing and In℡Ligent Interaction (ACII). https://doi.org/10.1109/ACII52823.2021.9597452
29.
HoM.-T.MantelloP.NguyenH.-K. T.VuongQ.-H. (2021). Affective computing scholarship and the rise of China: A view from 25 years of bibliometric data. Humanities and Social Sciences Communications, 8(1), 282. https://doi.org/10.1057/s41599-021-00959-8
30.
HollocksM. J.LerhJ. W.MagiatiI.Meiser-StedmanR.BrughaT. S. (2019). Anxiety and depression in adults with autism spectrum disorder: A systematic review and meta-analysis. Psychological Medicine, 49(4), 559–572. https://doi.org/10.1017/S0033291718002283
31.
HullL.PetridesK. V.AllisonC.SmithP.Baron-CohenS.LaiM.-C.MandyW. (2017). “Putting on my best normal”: Social Camouflaging in Adults with Autism Spectrum Conditions. Journal of Autism and Developmental Disorders, 47(8), 2519–2534. https://doi.org/10.1007/s10803-017-3166-5
32.
IencaM.MalgieriG. (2022). Mental data protection and the GDPR. Journal of Law and the Biosciences, 9(1), lsac006. https://doi.org/10.1093/jlb/lsac006
33.
The Japan Institute for Labour Policy and Training. (2020). Case 3: Exploring emotion recognition solutions to create daycare environments that are easier to work in: Kurara Daycare [Business Labor Trend]. https://www.jil.go.jp/kokunai/blt/backnumber/2020/11/018-023.pdf
34.
JáureguiD. A. G. (2017, June). Toward emotional internet of things for smart industry [Conference session]. SMART INTERFACES 2017, the Symposium for Empowering and Smart Interfaces in Engineering, Venice.
35.
KalantarianH.JedouiK.WashingtonP.TariqQ.DunlapK.SchwartzJ.WallD. P. (2019). Labeling images with facial emotion and the potential for pediatric healthcare. Artificial Intelligence in Medicine, 98, 77–86. https://doi.org/10.1016/j.artmed.2019.06.004
36.
KangE. B. (2023). On the praxes and politics of AI speech emotion recognition. In 2023 ACM Conference on Fairness, Accountability, and Transparency (pp. 455–466). Association for Computing Machineryhttps://doi.org/10.1145/3593013.3594011
37.
KatiraiA. (2023). Ethical considerations in emotion recognition technologies: A review of the literature. AI and Ethics. Advance online publication. https://doi.org/10.1007/s43681-023-00307-3
38.
KaurH.McDuffD.WilliamsA. C.TeevanJ.IqbalS. T. (2022). “I didn’t know I looked angry”: Characterizing observed emotion and reported affect at work. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems (pp. 1–18). Association for Computing Machinery.
39.
KimE.BryantD.SrikanthD.HowardA. (2021). Age bias in emotion detection: An analysis of facial emotion recognition performance on young, middle-aged, and older adults. In Proceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society (pp. 638–644). Association for Computing Machinery. https://doi.org/10.1145/3461702.3462609
40.
LandowskaA. (2019). Uncertainty in emotion recognition. Journal of Information Communication & Ethics in Society, 17(3), 273–291. https://doi.org/10.1108/JICES-03-2019-0034
41.
LevyK. (2022). Data driven: Truckers, technology, and the new workplace surveillance. Princeton University Press.
42.
LindsayS.OstenV.RezaiM.BuiS. (2021). Disclosure and workplace accommodations for people with autism: A systematic review. Disability and Rehabilitation, 43(5), 597–610.
43.
LordC.McCauleyJ. B.PepaL. A.HuertaM.PicklesA. (2020). Work, living, and the pursuit of happiness: Vocational and psychosocial outcomes for young adults with autism. Autism, 24(7), 1691–1703.
44.
MantelloP.HoM.-T. (2023). Emotional AI and the future of wellbeing in the post-pandemic workplace. AI & Society, 39, 1883–1889. https://doi.org/10.1007/s00146-023-01639-8
45.
MantelloP.HoM.-T.NguyenM.-H.VuongQ.-H. (2021). Bosses without a heart: Socio-demographic and cross-cultural determinants of attitude toward Emotional AI in the workplace. AI & Society, 38, 97–119. https://doi.org/10.1007/s00146-021-01290-1
46.
McStayA. (2018). Emotional AI: The rise of empathic media. Sage.
McStayA.UrquhartL. (2019). “This time with feeling?” Assessing EU data governance implications of out of home appraisal based emotional AI. First Monday, 24(10). https://doi.org/10.5210/fm.v24i10.9457
50.
MillerD.ReesJ.PearsonA. (2021). “Masking is life”: Experiences of masking in autistic and nonautistic adults. Autism in Adulthood, 3(4), 330–338. https://doi.org/10.1089/aut.2020.0083
51.
MohammadS. M. (2022). Ethics sheet for automatic emotion recognition and sentiment analysis (arXiv:2109.08256). arXiv. http://arxiv.org/abs/2109.08256
NagyJ. (2024). Autism and the making of emotion AI: Disability as resource for surveillance capitalism. New Media & Society, 26(7), 3989–4007. https://doi.org/10.1177/14614448221109550
56.
NakamuraK. (2019). My algorithms have determined you’re not human: AI-ML, reverse Turing-tests, and the disability experience. The 21st International ACM SIGACCESS Conference on Computers and Accessibility (pp. 1–2). Association for Computing Machinery. https://doi.org/10.1145/3308561.3353812
OlaL.Gullon-ScottF. (2020). Facial emotion recognition in autistic adult females correlates with alexithymia, not autism. Autism, 24(8), 2021–2034.
59.
O’NeilC. (2016). Weapons of math destruction: How big data increases inequality and threatens democracy. Crown.
60.
OngD. C. (2021, September28). An ethical framework for guiding the development of affectively-aware artificial intelligence [Conference session]. 2021 9th International Conference on Affective Computing and In℡Ligent Interaction (ACII), Nara, Japan. https://doi.org/10.1109/ACII52823.2021.9597441
61.
PasqualeF. (2020). New laws of robotics: Defending human expertise in the age of AI. The Belknap Press of Harvard University Press.
RoemmichK.SchaubF.AndalibiN. (2023). Emotion AI at work: Implications for workplace surveillance, emotional labor, and emotional privacy. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (pp. 1–20). Association for Computing Machinery. https://doi.org/10.1145/3544548.3580950
64.
RomualdezA. M.HeasmanB.WalkerZ.DaviesJ.RemingtonA. (2021). “People might understand me better”: Diagnostic disclosure experiences of autistic individuals in the workplace. Autism in Adulthood, 3(2), 157–167.
65.
SætraH. S. (2023). Technology and sustainable development: The promise and pitfalls of techno-solutionism. Taylor & Francis.
66.
SassonN. J.MorrisonK. E. (2019). First impressions of adults with autism improve with diagnostic disclosure and increased autism knowledge of peers. Autism, 23(1), 50–59. https://doi.org/10.1177/1362361317729526
67.
SedenbergE.ChuangJ. (2017). Smile for the camera: Privacy and policy implications of emotion AI. arXiv. arXiv:1709.00396.
68.
ShimoS. (2020). Risks of bias in AI-based emotional analysis technology from diversity perspectives. In 2020 IEEE International Symposium on Technology and Society (ISTAS) (pp. 66–68). IEEE. https://doi.org/10.1109/ISTAS50296.2020.9462168
69.
SmethurstL. J.ThompsonA. R.FreethM. (2024). “I’ve absolutely reached rock bottom and have no energy”: The lived experience of unemployed and underemployed autistic adults. Autism in Adulthood. Advance online publication. https://doi.org/10.1089/aut.2023.0171
70.
SoperR.BennetK.RivasP.Mathana. (2020). Developing Use cases to support an empathic technology ethics standard. In 2020 IEEE International Symposium on Technology and Society (ISTAS) (pp. 25–28). IEEE. https://doi.org/10.1109/ISTAS50296.2020.9462177
71.
StarkL.HoeyJ. (2021). The ethics of emotion in artificial intelligence systems. In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency (pp. 782–793). Association for Computing Machinery. https://doi.org/10.1145/3442188.3445939
72.
StarkL.StanhausA.AnthonyD. L. (2020). “I don’t want someone to watch me while I’m working”: Gendered views of facial recognition technology in workplace surveillance. Journal of the Association for Information Science and Technology, 71(9), 1074–1088.
73.
SteinertS.FriedrichO. (2020). Wired emotions: Ethical issues of affective brain-computer interfaces. Science and Engineering Ethics, 26(1), 351–367. https://doi.org/10.1007/s11948-019-00087-2
74.
TalaatF. M. (2023). Real-time facial emotion recognition system among children with autism based on deep learning and IoT. Neural Computing and Applications, 35(17), 12717–12728. https://doi.org/10.1007/s00521-023-08372-9
75.
TrevisanD. A.BoweringM.BirminghamE. (2016). Alexithymia, but not autism spectrum disorder, may be related to the production of emotional facial expressions. Molecular Autism, 7(1), 46. https://doi.org/10.1186/s13229-016-0108-6
Waisman-NitzanM.GalE.SchreuerN. (2021). “It’s like a ramp for a person in a wheelchair”: Workplace accessibility for employees with autism. Research in Developmental Disabilities, 114, 103959.
78.
WalkowiakE. (2021). Neurodiversity of the workforce and digital transformation: The case of inclusion of autistic workers at the workplace. Technological Forecasting and Social Change, 168, 120739. https://doi.org/10.1016/j.techfore.2021.120739
79.
WalkowiakE. (2023). Digitalization and inclusiveness of HRM practices: The example of neurodiversity initiatives. Human Resource Management Journal, 34, 1748–8583. https://doi.org/10.1111/1748-8583.12499
80.
WhelpleyC. E.BanksG. C.BochantinJ. E.SandovalR. (2021). Tensions on the spectrum: An inductive investigation of employee and manager experiences of autism. Journal of Business and Psychology, 36, 283–297.
