GenAI, Robots, and Inclusive Special Education: Autism Spectrum Disorder in the Age of Generative AI

Introduction

What if all our fictions would come true and become an integral part of our lives? This is exactly what happened in the context of artificial intelligence (AI) and robots. For centuries, from legends to stories, from myths to science fiction films, mankind has been in search of human-like intelligent elements and robots that behave like humans (Bozkurt, 2023a; Şenocak et al., 2023). This decades-long quest has several critical turning points in history. The three thousands-year-long quest, which started with René Descartes' (1637) introduction of the Automoto in his Discourse on the Method, continued with Karel Čapek's (1921) first use of the word 'robot' in the science fiction game Rossum's Universal Robots, and continued in 1955 when John McCarthy, Marvin Minsky, Nathaniel Rochester, and Claude Shannon (1955) used the term 'artificial intelligence' in The Dartmouth Summer Research Project on Artificial Intelligence. This fire, which was ignited by Alan Turing (1950) with the question “Can machines think?”, has influenced the whole world by the 2000s. About a century after the use of the concept of robot and three quarters of a century after the concept of AI was introduced, mankind's quest turned from fiction to reality.

OpenAI (2022) introduced ChatGPT at the end of 2022, when mankind, who likes to challenge himself with the possibilities offered by AI, thought that this quest had matured. Natural language processing (NLP) and large language models (LLMs) forged by Generative Pre-trained Transformers (GPT) technology leading to Generative Artificial Intelligence (GenAI) which was beyond our expectations and predictions. According to Bozkurt (2023b) GenAI has “the potential to master human language, one of the most complex and sophisticated technologies ever created, and to use human knowledge to identify patterns that may elude human perception. When properly trained, ChatGPT is a powerful tool that can learn, unlearn, and relearn language, allowing it to adapt and evolve with human communication needs” (p. 199).

The introduction of Generative AI (GenAI) has resonated throughout the educational landscape, including its application in special education, specifically for Autism Spectrum Disorder (ASD) (Ansari et al., 2023; Bahroun et al., 2023; Bozkurt, 2023c; İpek et al., 2023; Lambert & Stevens, 2023; Lo, 2023; Montenegro-Rueda et al., 2023; Pradana et al., 2023; Yu & Guo, 2023). The conversational capabilities of GenAI hold promise for various subareas of special education, offering numerous opportunities (Marino et al., 2023). In the context of ASD, GenAI is suggested as a technology for diversifying learning experiences and supporting the development of learners with a range of needs (Johnson et al., 2023). Additionally, some studies highlight the potential of LLMs and GPT variants in the early detection of ASD (Li et al., 2023; Mukherjee et al., 2023). Beyond the basic application of GenAI, there is also an exploration into enhancing its capacity with social robots. Bertacchini et al. (2023) argue that combining the power of AI with visual stimuli and interactive robotics offers a novel and promising approach to support individuals with Autism Spectrum Disorder (ASD) in their neurodevelopmental journey. This integration could be instrumental for assessment, monitoring, and treatment purposes, suggesting a forward-thinking direction in the use of technology for special education needs.