[Good Business] Human centricity in AI: Putting people before the machine

The decisive question is whether human beings, in their pursuit of AI-powered efficiency, will remain humane Human centricity is a key principle in AI ethics, emphasizing that technology should serve human development and well-being rather than merely functioning as data inputs. AI systems must be designed with consideration for their impact on people, ensuring fairness, accountability, and transparency, while also involving stakeholder participation in the development process. Responsible innovation in AI should enhance human capabilities and dignity, requiring organizations to conduct ethical impact assessments and be prepared to address any harms that arise from AI deployment. This is AI-generated. Read the article for full context. Report any errors. Whenever a new artificial intelligence application is introduced, the first question is usually: What can it do? Can it write faster, predict better or reduce costs? These are reasonable business questions. But responsible leaders must ask a more important question: What will this technology do to people? This is the heart of human centricity, one of the eight guiding principles in the Analytics and AI Association of the Philippines Code of Ethics for AI Professionals (AAP Code). The Code places AI within a national aspiration: technological progress should uplift the life of every Filipino while supporting a just and dignified society and the common good. Human centricity means that people are the purpose of technology, not inputs or data points. AI should serve human development and well-being. Human beings should not be reorganized to serve the requirements of machines, algorithms or technology vendors. This matters because AI can appear competent without understanding the people it affects. A system may rank job applicants yet unfairly screen out women, older workers or persons with disabilities. It may approve loans quickly while penalizing people whose lives do not fit data patterns. A generative AI tutor may produce answers while quietly weakening a student’s capacity to think, write, and exercise judgment. The AAP Code joins human centricity with fairness, safety, privacy, accountability, reliability, transparency and environmental sustainability. These principles must be understood together. An inaccurate medical recommendation cannot be called human-centered. Neither can an opaque hiring system, an intrusive surveillance tool or an educational application that encourages cognitive dependency. This explains why human oversight is not satisfied by putting a person at the end of an automated process to rubber-stamp the machine’s output. The person must have the competence, authority, information and time to question the system. Otherwise, “human in the loop” becomes ethical window dressing. Human judgment remains essential because social reality is more complex than the data used to represent it. Data come from the past and may carry its exclusions, injustices and distortions. Quantifiable indicators capture only part of a person’s situation, if at all. People possess dignity that cannot be reduced to a score, and they deserve to be heard when an AI-assisted decision affects their livelihood, education, health, freedom or access to services. Human centricity must therefore begin before procurement or programming. Leaders should ask: What human need are we addressing? Who defined that need? Who will benefit? Who may bear the risks? Were affected workers, customers, students or citizens consulted? Is AI necessary and proportionate, or are we adopting it because it is fashionable? During design and deployment, organizations must test not only technical accuracy but also consequences for different groups. They should protect personal data, explain consequential decisions in language people understand and provide accessible ways to question or appeal them. They must monitor unexpected harms after deployment rather than treating launch day as the end of ethical responsibility. The Code also asks AI professionals to be honest about system capabilities and limitations. This is important for generative AI, whose fluent language can create an illusion of knowledge and understanding. Calling probabilistic output “intelligent” does not make it wise or truthful. Professionals must resist exaggerated marketing and communicate uncertainty, error rates and appropriate uses. Human-centered AI should augment human capability rather than cause its surrender. In the workplace, this means using AI to relieve drudgery while investing in workers’ learning, judgment, and employability. In education, it means helping students think more deeply, not allowing machines to do the thinking they need to develop. In public service, it means improving access and responsiveness without turning citizens into cases processed by unchallengeable systems. None of this is anti-innovation. On the contrary, it is a call for better innovation. A technology that produces short-term gains while harming trust, dignity, capability or social inclusion is badly designed. Responsible innovation expands what people can become and contribute. Organizations can make human centricity practical by requiring ethical impact assessments, multidisciplinary review, meaningful stakeholder participation and clear assignment of accountability. They should measure human outcomes alongside speed, accuracy and savings. Leaders should be prepared to redesign, suspend or reject a system when its harms cannot be adequately controlled. Ultimately, there is no autonomous machine to praise for a just outcome or blame for an unjust one. People choose the goals, data, models, safeguards and conditions of use. People also decide whose welfare counts. The decisive question is whether human beings, in their pursuit of AI-powered efficiency, will remain humane. Our task is to exercise the wisdom and moral courage needed to ensure that technological power serves every Filipino, respects human dignity and advances the common good. That is human centricity—and it must be the center of responsible AI. – Rappler.com Dr. Benito Teehankee is a Full Professor of the Ramon V. del Rosario College of Business at De La Salle University. He is Chairman of the Responsible AI Council of the Analytics and AI Association of the Philippines and Co-chair of the Shared Prosperity Committee of the Management Association of the Philippines. Below are other Good Business columns on AI: