AI In Assistive Technology

1) Speech Recognition and Voice Assistants

• Tools: Google Assistant, Siri, Microsoft Cortana, and custom AI bots

• Use: Help students with mobility or visual impairments control devices, take notes, or access information hands-free

2) Text-to-Speech (TTS) and Speech-to-Text (STT)

• AI Function: Converts spoken language into written text and vice versa with high level of accuracy

• Use: Supports students with dyslexia, visual impairments, or writing difficulties

1) Computer Vision for the Visually Impaired

• Tools: Seeing AI (Microsoft), Be My Eyes (AI-powered).

• Use: Describes surroundings, reads text, identifies objects and people.

2) AI in Navigation Tools

• Use: GPS and computer vision help students with visual impairments navigate campuses safely.

• AI Function: AI interprets real-time video to detect obstacles and provide audio guidance.

1) Adaptive Learning Platforms

• AI Function: Analyzes student performance and adapts content in real-time.

• Use: Helps students with ADHD, autism, or learning disabilities by personalizing learning pace and presentation style

2) Predictive Analytics

• AI Function: Uses historical data and behaviour patterns to make predictions and provide trends

• Use: Identifies students at risk of falling behind and suggests interventions.

As artificial intelligence becomes increasingly integrated into assistive technologies, it brings not only powerful opportunities but also brings into account important ethical considerations. Many people have opposed the development and integration of AI based on these concerns.

The first major concern is data privacy. AI systems often heavily rely on collecting and analyzing personal data, such as speech patterns, behavioural habits, or emotional cues, to function effectively. For students, particularly those with disabilities, this raises concerns about how their data is stored, who has access to it, and how this data may be used beyond its intended purpose.

Another important concern is algorithmic bias. AI systems are trained on large data sets and as a result, this can introduce algorithmic bias based on the general trends in the data. This is prevalent in Large Language Models (LLMs) such as ChatGPT where studies have found that these AI's are biased against minority groups due to being trained on data from the internet. This can inadvertently reinforce existing inequalities rather than reduce them.

Furthermore, students may become overrelient on AI. While AI can support independence, there is a risk that students may become overly dependent on these tools. This can potentially limit the development of a student's own executive functioning skills. 

The final ethical concern is regarding the sustainability of AI and the impact it has on the Earth. AI's require large data centres and lots of electronic equipment to operate. These facilities use a lot of energy and water to keep the electronic equipment running, which is an increasing concern as AI is being further developed and integrated into technology.