AI and Digital Inclusion: An Opportunity for Everyone?
Explore how Artificial Intelligence can promote or hinder digital inclusion, creating equitable access to technology for every individual.
In the debate on artificial intelligence, risks are often discussed: technological unemployment, surveillance, algorithmic bias. But there is also another side of the coin, brighter and less explored: that of inclusion.
Because AI, if designed carefully and responsibly, can become a powerful tool for breaking down barriers and reducing inequalities. With the European Accessibility Act coming into force from June 28, 2025, digital accessibility becomes mandatory for many companies in Europe, making digital inclusion not only an ethical goal but a legal and concrete necessity.
The Problem of Digital Exclusion
For many people, accessing technology means participating in public life, learning, working, communicating. Yet, millions of individuals remain excluded from it: due to disabilities, age, social or cultural conditions.
According to the Council of the European Union, 27% of the EU population over 16 lives with some form of disability – that's 101 million people: 1 in 4 adults. For most of them, the digital world we use every day is not designed for their needs.
This is where artificial intelligence can make a difference. Not to replace the human, but to make them freer.
Voice Technologies: The Voice as a Bridge
Voice interfaces open new possibilities for those who cannot use a keyboard or touchscreen. Virtual assistants like those integrated into smartphones and home devices allow sending messages, searching for information, and adjusting the environment with voice commands.
For a person with a motor disability, this is not just convenient: it's autonomy.
The same applies to speech synthesis and recognition technologies, which help those with difficulties in reading or writing. Microsoft Seeing AI, for example, uses artificial intelligence to describe environments, objects, and texts through a smartphone camera. The app is now also available for Android after years of iOS exclusivity, making this technology accessible to millions more users.
Breaking Down Language Barriers
Another crucial area is the linguistic one. Neural network-based machine translations are becoming increasingly accurate. This means breaking down barriers between people who speak different languages, but also making educational, informational, and cultural content accessible to those who would otherwise be excluded.
It is an important step towards a more equitable global society.
Personalized Education for All
Education also benefits from these innovations. Personalized learning tools, automated tutoring, inclusive platforms: these are all solutions that can help students with different needs to express their full potential.
As we discussed in the article "Algorithmic Bias: AI and Invisible Discrimination", even the most advanced artificial intelligence can reflect pre-existing inequalities. But with a conscious approach, it can instead become a tool for inclusion.
The Microsoft Case: AI for Accessibility
Microsoft, through its "AI for Accessibility" program, has for years supported the development of inclusive technologies. The initiative funds projects ranging from cognitive support to visual and auditory accessibility, promoting an ethical and human use of artificial intelligence.
The program has already supported over 450 projects worldwide, demonstrating that technological innovation can truly be at the service of inclusion. The partnership with Be My Eyes represents a concrete example of how AI can drastically improve the quality of life for people with visual impairments.
The Regulatory Revolution: European Accessibility Act
From June 28, 2025, the European Accessibility Act (EAA) will radically change the digital accessibility landscape in Europe. The directive obliges companies to ensure that digital products and services are accessible to people with disabilities.
This means that millions of websites, apps, and digital services will have to comply with accessibility standards, opening new opportunities for AI integration in accessibility. As highlighted by the ASPHI Foundation, "to design an Accessible, Fair, and Inclusive Artificial Intelligence, it is necessary to start from three pillars: awareness, responsibility, and good design."
Concrete Examples of Success
Be My Eyes: This revolutionary app has transformed the lives of over 500,000 blind people worldwide. Born to connect volunteers with people needing visual assistance, Be My Eyes integrated AI with Be My AI, reducing average assistance time from 12 to 4 minutes and automatically resolving 90% of cases.
Microsoft's Seeing AI: An app that uses artificial intelligence to describe the surrounding world to blind or visually impaired people, reading texts, recognizing faces and objects.
Google Lookout: Automatically converts audio to text in real-time, helping deaf or hard-of-hearing people follow conversations and presentations in over 20 languages.
Voice Dream Writer: A text editor that supports people with dyslexia through voice feedback and intelligent suggestions.
AI as a Tool for Democratization
Artificial intelligence can democratize access to services and opportunities that would otherwise remain the privilege of a few. Think of AI-based legal assistants that can provide basic consultations to those who cannot afford a lawyer, or medical diagnostic systems that bring specialized expertise to remote areas.
Risks to Avoid
However, we must remain vigilant about the risks. As analyzed in our article on "Predictive Surveillance", the same technology that includes can also exclude if misused.
It is crucial that inclusive solutions respect privacy, maintain human control, and do not create new forms of technological dependency.
Towards an Inclusive Future
AI does not solve the world's problems by itself. But it can amplify good practices, support change, and make opportunities more accessible. It can serve as a bridge between those who have more and those who have less.
Provided it is designed with a sensitive, attentive, human gaze.
Digital inclusion is a concrete goal, not a utopia. And every time artificial intelligence is used to listen to those without a voice, to include those on the margins, to value differences instead of erasing them, then it becomes what it should be: a tool for everyone.
Key Points
- AI can break down barriers: Voice and automatic translation technologies make technology accessible to people with different abilities and linguistic backgrounds
- Education becomes personalized: Adaptive learning tools allow each student to learn according to their own pace and methods
- A cultural approach is needed: Inclusion must be designed from the very beginning, not added as an afterthought
- Results are already visible: Projects like Microsoft's AI for Accessibility demonstrate the concrete impact of inclusive technologies
FAQ – Frequently Asked Questions on AI and Digital Inclusion
Can AI really replace human assistance for people with disabilities? No, AI does not replace human assistance but complements it. Apps like Be My Eyes demonstrate that the best approach combines human volunteers and AI assistants, using artificial intelligence to reduce wait times (from 12 to 4 minutes) while maintaining the possibility of human assistance when necessary.
What are the main risks of AI for accessibility? The main risks include bias in training datasets, which can exclude people with specific disabilities (such as speech disorders), and excessive dependence on technology. As research highlights, many speech recognition systems do not work well for people with speech disabilities because the training data does not include representative samples.
Does the European Accessibility Act also apply to small businesses? Micro-enterprises with fewer than 10 employees are exempt from the obligation, but not from the ethical responsibility. Even for them, a discrimination complaint can have real consequences under Law 67/2006, which protects the rights of people with disabilities.
What AI technologies are already available for inclusion? Numerous solutions exist: Microsoft Seeing AI for visual description, Google Lookout for daily assistance, Be My Eyes for visual assistance via volunteers and AI, advanced speech synthesis systems, and automatic translation technologies to break down language barriers.
How can we ensure that inclusive AI is truly accessible? The direct involvement of people with disabilities in the design and testing of systems is crucial. According to the ASPHI Foundation, what is needed is "awareness, responsibility, and good design" with representative datasets and collaborative design that values the needs and desires of marginalized communities.