Topic: Multimodal Emotion Recognition and Responsible AI for Next-Generation Affective Computing

Affective computing, an interdisciplinary domain that bridges artificial intelligence (AI), psychology, neuroscience, cognitive science, and human-computer interaction, is undergoing a transformative evolution. The growing convergence of multimodal sensing technologies and intelligent computational modeling has unlocked unprecedented opportunities for machines to understand and respond to human affective states with greater nuance, empathy, and contextual awareness than previously possible.

Recent advances in computer vision, natural language processing, speech and audio signal analysis, wearable sensor data processing, and physiological signal interpretation have expanded the capabilities of affective systems. These technologies enable AI to detect subtle emotional cues from facial expressions, tone of voice, gestures, text semantics, and biometric responses, facilitating richer and more human-like interactions between humans and intelligent systems.

As we move toward the era of next-generation affective computing, it is crucial to pair technological progress with responsible and ethical AI principles. The ability of machines to sense and influence emotions raises profound questions about bias, fairness, transparency, privacy, and accountability. Developing emotion-aware AI systems that are trustworthy, explainable, and culturally inclusive is essential to ensure these technologies serve human well-being rather than compromise it.

This Special Issue seeks high-quality, original research contributions that push the frontiers of multimodal emotion recognition, sentiment understanding, and responsible affective AI frameworks. We encourage submissions that explore both technical innovations and societal dimensions, including:

● Advanced multimodal data fusion and representation learning methods;

● Emotion recognition incross-domain, cross-lingual, and low-resource settings;

● Bias detection and mitigation strategies in affective datasets and models;

● Transparent and interpretable models for emotion prediction;

● Ethical and privacy-preserving design in emotion-aware systems;

● Applications in mental health, education, social robotics, healthcare, marketing, and creative media.

By bringing together diverse perspectives and disciplines, this Special Issue aims to foster a holistic understanding of how multimodal affective intelligence can be developed and deployed responsibly, paving the way for emotionally intelligent, human-aligned, and ethically sound AI ecosystems.

Multimodal emotion recognition, affective computing, responsible AI, ethical human–AI interaction