مطالب

Objective: This review explores the trustworthiness of multimodal artificial intelligence (AI) systems, specifically focusing on vision-language tasks. It addresses critical challenges related to fairness, transparency, and ethical implications in these systems, providing a comparative analysis of key tasks such as Visual Question Answering (VQA), image captioning, and visual dialogue. Background: Multimodal models, particularly vision-language models, enhance artificial intelligence (AI) capabilities by integrating visual and textual data, mimicking human learning processes. Despite significant advancements, the trustworthiness of these models remains a crucial concern, particularly as AI systems increasingly confront issues regarding fairness, transparency, and ethics. Methods: This review examines research conducted from 2017 to 2024, focusing on forenamed core vision-language tasks. It employs a comparative approach to analyze these tasks through the lens of trustworthiness, underlining fairness, explainability, and ethics. This study synthesizes findings from recent literature to identify trends, challenges, and state-of-the-art solutions. Results: Several key findings were highlighted. Transparency: The explainability of vision language tasks is important for user trust. Techniques, such as attention maps and gradient-based methods, have successfully addressed this issue. Fairness: Bias mitigation in VQA and visual dialogue systems is essential for ensuring unbiased outcomes across diverse demographic groups. Ethical Implications: Addressing biases in multilingual models and ensuring ethical data handling is critical for the responsible deployment of vision-language systems. Conclusion: This study underscores the importance of integrating fairness, transparency, and ethical considerations in developing vision-language models within a unified framework.

Background: Throughout history, humans have consistently developed groundbreaking technologies, from fire and the wheel to modern computing, showcasing their ability to innovate and control their creations. The rise of artificial intelligence has sparked renewed concerns about whether machines might eventually surpass human intelligence and autonomy.Aims: This study aims to critically assess AI’s role in human society, particularly addressing concerns that it may surpass human control and agency. It seeks to demonstrate that while AI is a powerful tool, it lacks autonomy, self-augmentation, and intentionality, making it unlikely to replace human decision-making. Additionally, the paper examines historical technological advancements, showing how humans have always adapted and controlled new innovations.Methodology: Employing a historical-comparative methodology, this study traces the evolution of computing technologies from early tally systems to quantum computing. It incorporates philosophical analysis through the works of Hubert Dreyfus, John Searle, and Michael Tomasello, assessing AI’s limitations in replicating human cognition.Findings: The study finds that despite AI’s rapid advancements, it remains fundamentally dependent on human input, lacks true understanding, and is incapable of independent self-enhancement. The historical trajectory of technological progress demonstrates that while new technologies can disrupt societies, humans have consistently adapted and maintained control. Philosophical critiques of AI further reinforce the argument that intelligence is not solely computational but deeply rooted in embodiment, intuition, and shared intentionality—qualities that AI lacks.Conclusions: Concerns about AI overwhelming human agency are largely misplaced. Just as humanity has managed previous technological revolutions—including writing, mechanization, and computing—AI will be integrated and regulated according to human needs and ethical considerations. While vigilance is necessary, the myth of AI autonomy is exaggerated.