Shiho Arai

AI reshapes power, autonomy, and workplace culture in ways that technical frameworks alone cannot capture. Worker‑protection policies must evolve as fast as AI systems do—and they must reflect cultural, structural, and lived realities. The global majority will experience AI’s labor impact first and most intensely, which makes inclusive governance essential. Future of Work is not a technical debate. It is a social and cultural one. So what forms of power redistribution should AI policy prepare for?

The Hidden Danger of Emotion AI in the Workplace When I visited a client’s call center years ago, I felt a deep sense of discomfort—one that I couldn’t articulate at the time. Today, with the rise of AI-driven employee monitoring, I finally understand why. The company had introduced an AI system that “reads” emotions from the operator’s voice. If a customer complains, a human naturally feels frustration, sadness, or even anger. These emotions are part of being human. But the AI system labeled these emotions as indicators of “poor suitability” for the job. This is where the danger begins. Humans often hide their internal emotions during customer service. We regulate our expressions, maintain professionalism, and protect our inner world. But when a machine exposes those internal emotions—without context, without nuance— it crosses a boundary that should never be crossed. AI must never be allowed to evaluate the human inner life. Our internal emotions, our psychological states, our momentary reactions— these belong to a domain that AI should not invade. If AI begins labeling workers based on their internal emotional signals, we risk creating a future where: -employees are sorted and excluded based on momentary feelings -psychological privacy disappears -human dignity is compromised -HR decisions become detached from context and humanity AI can support human welfare—but only if we draw clear ethical boundaries. The inner life of a human being must remain off-limits. No algorithm should decide a person’s worth based on their emotions. This is the line we must protect.

AI adoption isn’t driven by technology alone — it’s driven by social exchange. After observing how organizations react to AI usage and token counts, I realized something: People don’t adopt AI based on capability. They adopt it based on perceived fairness, reciprocity, and trust — exactly what Social Exchange Theory explains. In my opinion, in most organizations AI adoption is framed as an individual contribution, not an organizational norm. Instead, we can ask ourselves • Will my AI‑assisted work be judged fairly? • Does my manager reward AI‑driven output the same way as human effort? • Is AI increasing my value — NOT replacing it? When these questions aren’t answered, AI usage drops. Not because AI is weak, but because the social exchange feels unbalanced. I’m curious to hear from others: In your organization, is AI cost and impact clearly communicated across your teams? #AcademicLinkedIn #AImaturity #OrganizationalCulture #AIadoption #SocialExchangeTheory #FutureOfWork #AIandSociety #HumanCenteredAI

Work‑From‑Home AI Opportunity（Flexible & Global） Want to break into AI training or data annotation? Outlier is looking for new remote experts for multiple projects — AI evaluation, safety review, linguistic tasks, transcription, and more. Apply here: https://lnkd.in/gVxdKDFZ Why Outlier? • Fully remote • Competitive pay • Flexible hours • Roles for specialists (legal, psychology, medical, linguistics) If you sign up through this link and get hired, you’ll gain access to Outlier’s full project pool. Note: I’m only sharing the link. For questions, please check Outlier’s official page. What is Outlier? Outlier works with top tech firms, offer flexible remote roles, and give experts access to diverse projects—from safety review to linguistic evaluation and specialized domains like legal, psychology, and medical AI.