The Impact of AI-Powered Smart Home Devices on Individuals with IDD Through Enhanced Autonomy

By Shezin Waziha Hussain

1.1 AI implementation in smart home devices and its impact

A smart home is a residence equipped with technology that allows for the automation and 

remote control of various household systems and devices, such as lighting, climate control, appliances, security, and entertainment. These devices can be managed through a central hub or individually, either via a wired or wireless network. For individuals with IDD, smart technology offers enhanced accessibility, such as voice-activated doors and appliances, simplifying daily tasks. Smart homes promote independence by reducing the reliance on others for task completion.  Additionally, these systems can alert caregivers during emergencies, help prevent accidents or falls through reminders, and schedule tasks like medication intake, contributing to overall safety and well-being (Paz, 2023).

Jamwal et al (2022) highlight that smart home communication technology can improve outcomes for people living with disabilities and complex needs especially, in areas like independence, participation, and quality of life. Several factors impact the successful implementation of technology such as personalization of technology to individual needs, flexibility of technology systems, and ongoing support provided to the person with disability and their close others. Jamwal et al., (2022) also mentions that ethical considerations need to be taken into account when implementing technology for this vulnerable population, including – a) potential risks through increased social participation, and b) privacy concerns related to monitoring technologies. The paper highlights the limitations mentioning that there is a lack of high-quality study design and limited use of established outcome measures relevant to this population. Future research should focus on utilizing more rigorous study designs, using validated outcome measures specific to people with disabilities, and evaluating technology implementation in real-world settings rather than just pilot testing. This paper also highlights the potential benefits of smart homes and communication technology for people with disabilities, while also emphasizing the need for more robust evidence to guide implementation and policy decisions in this rapidly evolving field.

2.2 AI Technology and its Role in Supporting Autonomy for Individuals with Intellectual and Developmental Disabilities (IDD)

Respect for autonomy is the first of four key ethical principles of the European Commission's high-level expert group's ethics guidelines for trustworthy AI (Ethics Guidelines for Trustworthy AI | Shaping Europe’s Digital Future, 2019). As artificial systems continue to take on a growing range of tasks, policymakers are under increasing pressure to tackle the emerging risks associated with their use (Prunkl, 2024). Although advancements have been made in various areas, the existing policy literature on autonomy and AI remains fragmented, consisting of a collection of seemingly unrelated recommendations aimed at addressing a diverse range of potential risks (Prunkl, 2024).

Cognitive impairment, also known as intellectual disability (ID), is a condition that impacts an individual's capacity to learn and carry out everyday tasks. This disorder can stem from a variety of sources, including congenital abnormalities, infections, hereditary factors, and traumatic events. Young people affected by ID often face difficulties in areas such as verbal communication, mobility, self-care in dressing, and independent eating. They may also experience challenges in interpreting social signals and forming friendships. It's worth noting that intellectual disability could be associated with one or multiple elements depicted in the referenced visual representation (Almufareh et al., 2023).

AI-powered live captioning systems convert spoken words into text in real time, offering significant advantages to those with hearing impairments or deafness. This technology has wide-ranging applications, proving valuable in diverse environments such as educational institutions, professional gatherings, and musical performances (Kawas et al., 2016; Millett, 2021).

In educational settings, Millett (2021) in her study on speech-to-text captioning accuracy for deaf and hard-of-hearing students, evaluated the accuracy of 8 different speech-to-text apps and software platforms. It tested captioning accuracy for a university lecture, a video of the lecture, and a conversation between 3 students, all under controlled audio conditions. For the live lecture, 4 out of 5 technologies exceeded 90% accuracy, with Google Slides and Otter achieving 98-99% accuracy. Video captioning had the highest overall accuracy, with 5 out of 6 technologies achieving over 90% accuracy. YouTube, Microsoft Stream, and Otter reached 98-99% accuracy. For captioning a real-time conversation between 3 students, both technologies tested (Ava and Microsoft Translator) exceeded 90% accuracy. The study concluded that, given excellent audio quality, speech-to-text technology accuracy is sufficient to consider use by postsecondary students. However, the author notes that individual student needs and abilities should be considered when evaluating the effectiveness of captioning as an accommodation.

As demonstrated in the research by Camgoz et al. (2018) and Guo et al. (2018), technology for interpreting sign language can transform manual gestures into written text or spoken words. This innovation bridges the communication gap between individuals who are deaf or hard of hearing and those unfamiliar with sign language. Such advancements facilitate more inclusive and accessible interactions across diverse populations.

Bhattacharjee et al., (2020) in their research, they highlight that providing autonomous assistance for activities of daily living (ADL) has been a long-standing objective in robotics, with a focus on tasks like food handling and feeding (Gallenberger et al., 2019), personal hygiene (Hawkins et al., 2012), retrieving objects from the floor (Choi et al., 2009), and passing items to others (Strabala et al., 2013). The key highlights from Bhattacharjee et al., (2020) mentions that user preferences for different levels of robot autonomy in assisted feeding, considering the trade-offs between autonomy, potential errors, and required user input.

Although significant progress has been made toward sustainable solutions, fully autonomous systems are still not ready for widespread use due to the frequency and seriousness of errors that automation can introduce. While complete autonomy may be years away, many users could benefit from partial solutions. Semi-autonomous systems, which balance autonomy with increased user control, offer one potential way to address the challenges of automation errors.

Organizational autonomy can also play a role for people with intellectual disabilities. Estreder et al. (2024) in their work mentions about inclusive environments for workers with intellectual disabilities. Autonomy support from supervisors enhances workplace well-being. Non-disabled supervisors should provide autonomy to support workers. Training in social skills for supervisors is recommended. Supervisors should avoid categorizing workers with disabilities as excluded groups and should promote practices that encourage autonomy satisfaction in the workplace.

1.3 Considerations for selecting, obtaining, and training AI-powered home devices for individuals with IDD  

Machine learning is one of the parts of artificial intelligence, when algorithms are exposed to more data, they can learn and improve from it to anticipate consumers’ needs. AI can remove accessibility barriers through different solutions such as image recognition and facial recognition for people with visual impairment, lip reading recognition for people with a hearing impairment, text summarization for people with mental impairment, and real-time captioning or translation for people with learning impairment or even people who don't speak the language (Martinez, 2021). Martinez (2021) mentions further that artificial intelligence significantly transforms the daily lives of individuals with disabilities. For example, a person with cognitive impairment can better understand their surroundings through tools like text summarization, which simplifies complex messages into easily understandable content. Tasks that were once challenging or unattainable have become accessible and manageable. AI creates a more inclusive world by adapting to the needs of individuals with disabilities ensuring their challenges are acknowledged and addressed. This technological advancement promotes equality by leveling the playing field, and everyone regardless of ability will be able to engage and participate on equal terms.

1.4 Outcomes of implementation  

According to the Office of the High Commissioner for Human Rights (OHCHR) (2021), the report examines how AI technologies affect the rights of persons with disabilities, both positively and negatively, thus emphasizing the need for a human rights-based approach to AI development and deployment. AI has the potential to enhance accessibility, improve healthcare, and promote independent living for persons with disabilities via the usage of assistive technology, early diagnosis of conditions, and personalized learning tools.

The report also provides several key recommendations regarding AI and its impact on people with disabilities. From the aspect of state obligations, states should ensure that AI systems respect protect and fulfill the rights of persons with disabilities inclusive of implementing legal and policy frameworks to prevent discrimination, ensuring accessibility of AI systems, and providing effective remedies for rights violations. Regarding participation and consultation, state and private actors should actively involve disability organizations in AI-related decision-making and ensure meaningful participation of persons with disabilities in AI design and implementation. That report also recommends inclusivity by adopting universal design principles in AI development, ensuring AI systems are accessible to persons with various disabilities, and providing reasonable accommodations where necessary. The report highlights the need to implement robust data protection measures for persons with disabilities, ensure informed consent in data collection and use, and protect against privacy violations and data misuse. Lastly, the report calls for establishing mechanisms to monitor AI impacts on disability rights, ensuring accountability for rights violations related to AI systems, and conducting human rights impact assessments of AI technologies.

Conclusion

Smart technology is gaining widespread popularity, providing individuals with

disabilities the opportunity to live independently and comfortably, even when living alone.

Devices like smart lighting, garage openers, and thermostats can be managed remotely using smartphones, tablets, or voice commands. These innovations empower an individual with limited mobility to control their home environment more efficiently, improving their daily lives. 

Smart home and communication technology use impacts various stakeholders, including individuals with disabilities, their informal caregivers or close family members, paid support workers, and healthcare professionals. While these technologies can enhance independence and participation for people with disabilities, each stakeholder group must evaluate the resources they require or can provide to ensure these tools' continued and effective use (Jamwal et al., 2022). 

Artificial Intelligence plays a key role in the creation of assistive robots, self-driving wheelchairs, and navigation systems designed for individuals with visual impairments (Kahraman & Turhan, 2022; Kim et al., 2023; Lim et al., 2023). AI-driven smart home systems have been demonstrated to enhance independent living skills, improve quality of life, and support psychological well-being (Landuran et al., 2023). 

Ethics and privacy play a role in smart device usage by people with intellectual disabilities. Hine et al., (2022) in their work highlight the fact that besides people with Intellectual disabilities and their use of smart devices, healthcare professionals may encounter challenges in navigating ethical practices while also acting as advocates for the technology. They often take the role of educating and reassuring users who may have limited technical knowledge of the systems. This highlights the importance of conducting research into the lived experiences are smart care users alongside the technical and clinical studies that drive implementation. Such research should inform the design process, contribute to more robust preemptive ethical reviews, and support the training and guidance of healthcare professionals involved in deploying these systems. 

Privacy seems to be a concerning issue for users of smart devices particularly for people with disabilities. Individuals with IDD experience challenges in intellectual functioning, which manifest as difficulties in processing, learning, and retaining information (Chalghoumi, 2012). When people with disabilities use IT, it is adapted or utilized to address their challenges, making it not a matter of choice, luxury, or privilege. Instead, it becomes a standard form of support, similar to how technology is used in everyday life by the general population (Dawe, 2006; Kerkhof et al., 2016). Privacy and security risks associated with its use are rapidly increasing, leaving individuals with IDD particularly susceptible to breaches. Many individuals with IDD may struggle to fully comprehend the risks and benefits of using IT and often face challenges due to limited literacy skills (Debatin et al., 2009; Iachello & Hong, 2007). Challenges in processing information and understanding abstract or complex concepts can make individuals with IDT more susceptible to privacy risks, including cyberbullying, financial exploitation, and sexual abuse (Didden et al., 2009; Holmes & O’Loughlin, 2014; Löfgren-Mårtenson, 2008). 

Despite this, research and technological advancements have frequently neglected the ethical challenges individuals with IDD encounter when using technology, focusing instead on highlighting the benefits of IT for this group. Additionally, there are significant limitations of prior research showing there is a tendency to collect information about individuals with IDD indirectly, relying on caregivers or third parties, therefore conducting research about them rather than involving them directly (Chalghoumi et al., 2019). Privacy behaviors encompass an individual's past, present, and anticipated responses to perceived privacy risks. Gaining insight into these attitudes and actions is essential for proactively designing and implementing effective privacy practices, policies, and regulations, as well as for developing technologies that prioritize and respect users' privacy. (Debatin et al., 2009; Iachello & Hong, 2007; Kokolakis, 2017). 

References

Almufareh, M. F., Tehsin, S., Humayun, M., & Kausar, S. (2023). Intellectual Disability and Technology: An Artificial Intelligence Perspective and Framework. Journal of Disability Research, 2, 58–70. https://doi.org/10.57197/JDR-2023-0055

Bhattacharjee, T., Gordon, E. K., Scalise, R., Cabrera, M. E., Caspi, A., Cakmak, M., & Srinivasa, S. S. (2020). Is More Autonomy Always Better?: Exploring Preferences of Users with Mobility Impairments in Robot-assisted Feeding. Proceedings of the 2020 ACM/IEEE International Conference on Human-Robot Interaction, 181–190. https://doi.org/10.1145/3319502.3374818

Camgoz, N. C., Hadfield, S., Koller, O., Ney, H., & Bowden, R. (2018). Neural Sign Language Translation. 7784–7793. https://openaccess.thecvf.com/content_cvpr_2018/html/Camgoz_Neural_Sign_Language_CVPR_2018_paper.html

Chalghoumi, H. (2012). Rethinking the definition of assistive technology in education. Canadian Journal of Learning and Technology / La Revue Canadienne de l’apprentissage et de La Technologie, 38(3). https://www.learntechlib.org/p/178045/

Chalghoumi, H., Cobigo, V., Dignard, C., Gauthier-Beaupré, A., Jutai, J. W., Lachapelle, Y., Lake, J., Mcheimech, R., & Perrin, M. (2019). Information Privacy for Technology Users With Intellectual and Developmental Disabilities: Why Does It Matter? Ethics & Behavior, 29(3), 201–217. https://doi.org/10.1080/10508422.2017.1393340

Choi, Y. S., Deyle, T., Chen, T., Glass, J. D., & Kemp, C. C. (2009). A list of household objects for robotic retrieval prioritized by people with ALS. 2009 IEEE International Conference on Rehabilitation Robotics, 510–517. https://doi.org/10.1109/ICORR.2009.5209484

Dawe, M. (2006). Desperately seeking simplicity: How young adults with cognitive disabilities and their families adopt assistive technologies. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 1143–1152. https://doi.org/10.1145/1124772.1124943

Debatin, B., Lovejoy, J. P., Horn, A.-K., & Hughes, B. N. (2009). Facebook and Online Privacy: Attitudes, Behaviors, and Unintended Consequences. Journal of Computer-Mediated Communication, 15(1), 83–108. https://doi.org/10.1111/j.1083-6101.2009.01494.x

Didden, R., Scholte, R. H. J., Korzilius, H., De Moor, J. M. H., Vermeulen, A., O’Reilly, M., Lang, R., & Lancioni, G. E. (2009). Cyberbullying among students with intellectual and developmental disability in special education settings. Developmental Neurorehabilitation, 12(3), 146–151. https://doi.org/10.1080/17518420902971356

Estreder, Y., Martínez-Tur, V., Moliner, C., Gracia, E., & Lira, E. (2024). The voice of persons with intellectual disability: Why does autonomy support make them feel better? Applied Psychology, 73(3), 1055–1070. https://doi.org/10.1111/apps.12487

Ethics guidelines for trustworthy AI | Shaping Europe’s digital future. (2019). https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai

Gallenberger, D., Bhattacharjee, T., Kim, Y., & Srinivasa, S. S. (2019). Transfer Depends on Acquisition: Analyzing Manipulation Strategies for Robotic Feeding. 2019 14th ACM/IEEE International Conference on Human-Robot Interaction (HRI), 267–276. https://doi.org/10.1109/HRI.2019.8673309

Guo, D., Zhou, W., Li, H., & Wang, M. (2018). Hierarchical LSTM for Sign Language Translation. Proceedings of the AAAI Conference on Artificial Intelligence, 32(1), Article 1. https://doi.org/10.1609/aaai.v32i1.12235

Hawkins, K. P., King, C.-H., Chen, T. L., & Kemp, C. C. (2012). Informing assistive robots with models of contact forces from able-bodied face wiping and shaving. 2012 IEEE RO-MAN: The 21st IEEE International Symposium on Robot and Human Interactive Communication, 251–258. https://doi.org/10.1109/ROMAN.2012.6343762

Hine, C., Nilforooshan, R., & Barnaghi, P. (2022). Ethical considerations in design and implementation of home-based smart care for dementia. Nursing Ethics, 29(4), 1035–1046. https://doi.org/10.1177/09697330211062980

Holmes, K. M., & O’Loughlin, N. (2014). The experiences of people with learning disabilities on social networking sites. British Journal of Learning Disabilities, 42(1), 1–5. https://doi.org/10.1111/bld.12001

Iachello, G., & Hong, J. (2007). End-User Privacy in Human–Computer Interaction. Foundations and Trends® in Human–Computer Interaction, 1(1), 1–137. https://doi.org/10.1561/1100000004

Jamwal, R., Jarman, H. K., Roseingrave, E., Douglas, J., & Winkler, D. (2022). Smart home and communication technology for people with disability: A scoping review. Disability and Rehabilitation: Assistive Technology, 17(6), 624–644. https://doi.org/10.1080/17483107.2020.1818138

Kahraman, M., & Turhan, C. (2022). An intelligent indoor guidance and navigation system for the visually impaired. Assistive Technology, 34(4), 478–486. https://doi.org/10.1080/10400435.2021.1872738

Kawas, S., Karalis, G., Wen, T., & Ladner, R. E. (2016). Improving Real-Time Captioning Experiences for Deaf and Hard of Hearing Students. Proceedings of the 18th International ACM SIGACCESS Conference on Computers and Accessibility, 15–23. https://doi.org/10.1145/2982142.2982164

Kerkhof, Y. J. F., Graff, M. J. L., Bergsma, A., De Vocht, H. H. M., & Dröes, R.-M. (2016). Better self-management and meaningful activities thanks to tablets? Development of a person-centered program to support people with mild dementia and their carers through use of hand-held touch screen devices. International Psychogeriatrics, 28(11), 1917–1929. https://doi.org/10.1017/S1041610216001071

Kim, S. K., Jang, J.-W., Hwang, Y. S., Lee, O. E., & Jo, H. S. (2023). Investigating the effectiveness of Socially Assistive Robot on Depression and Cognitive Functions of Community Dwelling Older Adults with Cognitive Impairments. Assistive Technology, 0(0), 1–9. https://doi.org/10.1080/10400435.2023.2237554

Kokolakis, S. (2017). Privacy attitudes and privacy behaviour: A review of current research on the privacy paradox phenomenon. Computers & Security, 64, 122–134. https://doi.org/10.1016/j.cose.2015.07.002

Landuran, A., Sauzéon, H., Consel, C., & N’Kaoua, B. (2023). Evaluation of a smart home platform for adults with Down syndrome. Assistive Technology, 35(4), 347–357. https://doi.org/10.1080/10400435.2022.2075487

Lim, M.-J., Song, W.-K., Kweon, H., & Ro, E.-R. (2023). Care robot research and development plan for disability and aged care in Korea: A mixed-methods user participation study. Assistive Technology, 35(4), 292–301. https://doi.org/10.1080/10400435.2022.2038307

Löfgren-Mårtenson, L. (2008). Love in Cyberspace: Swedish Young People with Intellectual Disabilities and the Internet1. Scandinavian Journal of Disability Research, 10(2), 125–138. https://doi.org/10.1080/15017410701758005

Martinez, C. (2021, March 5). Artificial Intelligence Enhances Accessibility for People with Disabilities. Inclusive City Maker. https://www.inclusivecitymaker.com/artificial-intelligence-accessibility-examples-technology-serves-people-disabilities/

Millett, P. (2021). Accuracy of Speech-to-Text Captioning for Students Who are Deaf or Hard of Hearing. 25, 1–13.

Paz, Z. (2023, January 6). Smart Home Technology: A Game-Changer for Disability Support and Care - LDRFA. https://www.ldrfa.org/smart-home-technology-people-disabilities/

Prunkl, C. (2024). Human Autonomy at Risk? An Analysis of the Challenges from AI. Minds and Machines, 34(3), 26. https://doi.org/10.1007/s11023-024-09665-1

Strabala, K. W., Lee, M. K., Dragan, A. D., Forlizzi, J. L., Srinivasa, S., Cakmak, M., & Micelli, V. (2013). Towards Seamless Human-Robot Handovers. Journal of Human-Robot Interaction, 2(1), 112–132. https://doi.org/10.5898/JHRI.2.1.Strabala