Sam Sepah on the High Cost of Accessibility: Disabled Users Forced to Sacrifice Privacy in the AI Age

“The internet is where anyone can access it,” Vint Cerf, a founding father of the internet, once proclaimed. Yet, a chilling truth echoes in his words. While the internet has opened doors for many, it has also demanded a high price of admission — the surrender of privacy and personal data. This raises concerns about the true cost of access, particularly for marginalized users, such as those with disabilities or who are Deaf. The very tools designed to enhance accessibility often require a sacrifice of privacy that non-disabled users might not face.

In the Western world, an estimated 350 million individuals with disabilities, encompassing physical, sensory, cognitive, or mental health conditions that impact their daily functioning, rely on assistive technologies to overcome challenges and improve their quality of life. In recent years, advancements in artificial intelligence (AI) have brought about exciting new technologies that offer increased independence and accessibility for people with disabilities. AI-powered cameras, equipped with sophisticated object recognition and image analysis capabilities, are being integrated into a range of assistive tools and technologies. These smart cameras can help users with disabilities navigate their surroundings, identify objects or voices, and perform tasks that were once difficult or impossible.

AI-powered assistive apps are revolutionizing accessibility for the blind, offering unprecedented independence through features like describing surroundings and deciphering text. However, this comes at a significant privacy cost. These apps constantly collect a stream of personal data, including where users go, who they meet, and even their most private activities at home including their medical information. For blind users, opting out of these apps isn’t a realistic option, as they dramatically improve their quality of life. While data collection is necessary for these apps to function, it highlights a critical trade-off: blind and low-vision users gain immense value from real-time assistance, but at the expense of their privacy.

Similarly, Deaf or hard-of-hearing users initially find empowerment and connection through speech-to-text transcription apps, overcoming communication barriers in appointments and family gatherings. The app transcribes spoken words into text in real-time, allowing them to participate fully and connect with others. However, they may be unaware of the app’s potential for extensive data collection, including recording other conversations without their knowledge or consent.

While many companies protect user data by processing it on-device, minimizing storage, offering opt-in data sharing, encrypting transmissions, and providing transparent privacy policies, the line between empowerment and exploitation can be razor-thin. The price of ignorance may be a huge risk, simply trusting the companies that provide these apps.

This exchange raises important questions: How long should this private data be stored? Who truly owns it? Is this data being shared with third parties, such as advertisers or pharmaceutical manufacturers? Is the trade-off for a “free” app justified? This vulnerability is further amplified when using apps from smaller startups, which often lack robust data privacy policies and the legal accountability that larger corporations have. This underscores the need for greater transparency and user control over their data.

A well-known data breach case, the 23andMe data breach serves as a cautionary tale for the handling of sensitive personal data. In 2018, this breach exposed customers’ sensitive personal information, including names, contact details, genetic data, and family history information to unauthorized third parties. Customers sued 23andMe, alleging that the company failed to adequately protect their data. The breach exposed sensitive information to potentially dangerous individuals or groups, raising concerns about identity theft, discrimination, and other forms of harm for affected customers, including those with disabilities who may have their disability-related information exposed on the dark web. This incident reminds us all of the need for strong data security, transparency, user control, and corporate accountability to ensure the privacy and safety of user information, particularly for those with disabilities.

The current landscape requires that disabled users disclose a wealth of personal data in order to utilize accessibility technologies. For instance, screen readers, which convert on-screen text to speech for visually impaired users, may require access to browsing history, search queries, and even sensitive information such as personal confidential information recorded to optimize performance and personalization. Without screen readers, this information would remain inaccessible to visually impaired users, forcing them to choose between privacy and access. Non-disabled users do not face this difficult choice. This exchange of privacy for accessibility presents a significant ethical dilemma.

Disabled users often find themselves in a precarious position, forced to weigh the benefits of accessibility against the risks to their privacy. The lack of robust privacy regulations and transparent data practices exacerbates this dilemma. While some companies may claim to anonymize or aggregate user data, the reality is that even de-identified information can be re-identified with sophisticated algorithms, leaving disabled users vulnerable.

It’s crucial for all users to be informed about data privacy practices and choose apps from reputable developers prioritizing user protection. It is important to understand where this data is stored, who has access to it, and how it is being protected. Users should also inquire about how this data is being used. Is it being used solely to provide accessibility features, or is it being used for other purposes, such as marketing or research?

On one hand, access to technology is fundamental for disabled individuals to engage in education, employment, and social interaction. With responsible usage and transparent policies, these apps can significantly enhance the lives of individuals with disabilities while safeguarding their fundamental rights. It is imperative that accessibility features are built with privacy by design principles, minimizing the collection and retention of personal data. Additionally, clear and concise privacy policies should be readily available, empowering disabled users to make informed choices about their data. The path forward must prioritize the empowerment of disabled users. They should not be forced to choose between accessibility and privacy. By championing privacy-centric accessibility solutions, we can create a digital landscape that is both inclusive and respectful of individual rights, ensuring that everyone can fully participate in the digital world without compromising their privacy.

About Sam Sepah

Deaf visionary Sam Sepah transforms organizations with inclusive leadership, AI research, and strategic community partnerships. He influences corporate and STEM leaders, fosters thriving teams, and champions user-centered design, especially for those with Deafness and disabilities. A bridge between industries and communities, Sam’s leadership shapes a more equitable future, highly recognized by both Deaf communities and Fortune 500 companies. He explores the world with his wife and three CODA daughters, embracing cultural experiences and creating lasting memories.