Aligning AI Development With Societal Values

"When only a handful of actors define how AI systems are built and used, public oversight erodes. These systems increasingly reflect the values and economic incentives of their creators, often at the expense of inclusion, accountability and democratic oversight. Without intervention, these trends risk entrenching structural inequities and shrinking the space for alternative approaches. This white paper outlines a strategic countervision: Public AI. It proposes a model of AI development and deployment grounded in transparency, democratic governance and open access to critical infrastructure. Public AI refers to systems that are accountable to the public, where foundational resources such as compute, data and models are openly accessible and every initiative serves a clearly defined public purpose. Grounded in a realistic analysis of the constraints across the AI stack – compute, data and models – the paper translates the concept of Public AI into a concrete policy framework with actionable steps. Central to this framework is the conviction that public AI strategies must ensure the continued availability of at least one fully open-source model with capabilities approaching those of proprietary state-of-theart systems. Achieving this goal requires three key actions: coordinated investing in the open-source ecosystem, providing public compute infrastructure, and building a robust talent base and institutional capacity. It calls for the continued existence of at least one fully open-source model near the frontier of capability and lays out three imperatives to achieve this: strengthening open-source ecosystems, investing in public compute infrastructure, and building the talent base to develop and use open models. To guide implementation, the paper introduces the concept of a “gradient of publicness” to AI policy – a tool for assessing and shaping AI initiatives based on their openness, governance structures, and alignment with public values. This framework enables policymakers to evaluate where a given initiative falls on the spectrum from private to public and to identify actionable steps to increase public benefit"

Humanizing AI Through the Kano Model In an era where generative AI has become a ubiquitous offering, true differentiation lies not in merely adopting the technology but in integrating human values into its core. Building on my earlier discussion about applying the Kano Model to Gen AI strategy, let’s explore how this framework can refocus development metrics to prioritize ethics and human-centricity. By aligning AI systems with human needs, organizations can shift from functional tools to trusted partners that inspire lasting loyalty. Traditional metrics such as speed, scalability, and model accuracy have evolved into basic expectations the “must-haves” of AI. What truly elevates a product today is its ability to embody values like safety, helpfulness, dignity, and harmlessness. These qualities, categorized as “delighters” in the Kano Model, transform AI from a transactional tool into a meaningful collaborator. Key Human-Centric Differentiators Safety: Proactive safeguards must ensure AI systems protect users from risks, whether physical, emotional, or societal. Safety is non-negotiable in building trust. Helpfulness: Personalized, context-aware interactions demonstrate empathy. AI should anticipate needs and adapt to individual preferences, turning routine tasks into meaningful experiences. Dignity: Ethical design principles—fairness, transparency, and privacy—must underpin AI development. Respecting user autonomy fosters long-term trust and engagement. Harmlessness: AI outputs and recommendations should prioritize user well-being, avoiding unintended consequences like bias, misinformation, or psychological harm. This human-centered approach represents a paradigm shift in technology development. While traditional KPIs remain important, they are no longer sufficient to stand out in a crowded market. Organizations that embed human values into their AI systems will not only meet user expectations but exceed them, creating emotional connections that drive loyalty. By applying the Kano Model, businesses can systematically align innovation with ethics, ensuring technology serves humanity rather than the other way around. The future of AI isn’t just about efficiency it’s about elevating human potential through thoughtful, responsible design. How is your organization balancing technical excellence with human values?

How do we scale Generative AI without compromising ethics, sustainability, or data integrity? Here are my ten principles: 🔹 Strong Data Foundation: Ensure clean, reliable, and well-structured data to build effective AI systems. 🔹 Bias Mitigation: AI must fairly represent all voices through diverse datasets and rigorous testing. 🔹 Energy Efficiency: Consider the full environmental footprint—carbon, water, and energy consumption—to minimize AI’s impact. 🔹 Transparency: Explainable AI is key to earning user trust by making decisions understandable. 🔹 Data Privacy: Privacy-first design must be prioritized to respect users’ growing data concerns. 🔹 Human Oversight: AI should enhance human judgment, with human-in-the-loop systems ensuring responsible outcomes. 🔹 Guardrails: Implement ethical guardrails to prevent misuse and ensure AI aligns with societal values. 🔹 Collaboration with Regulators: Work closely with regulators like the EU AI Act to ensure compliance and trust. 🔹 Continuous Monitoring and Auditing: Regularly audit AI systems to catch biases and inefficiencies, ensuring ongoing alignment with ethical goals. 🔹 Inclusive Development: Diverse, inclusive teams bring varied perspectives, helping avoid blind spots and foster fair AI. These principles offer a roadmap for scaling AI that is both innovative and responsible, ensuring a balance between growth and ethical standards. #ai #generativeai #responsibleai #genai #ethicalai

✨ AI at a crossroads: Can we steer it responsibly? The Association for the Advancement of Artificial Intelligence (AAAI) 2025 Presidential Panel on the Future of AI Research lays out a stark reality—AI is advancing at an unprecedented pace, but governance, safety, and evaluation mechanisms are struggling to keep up. 🌏 Having worked at the intersection of AI governance, responsible deployment, and multi-agent AI, I see a recurring challenge: we are building AI that is more powerful than our ability to govern it responsibly. 🔬 Key takeaways from the report & my perspective:- ✅ AI Reasoning & Trustworthiness:- While LLMs and Agentic AI are demonstrating emergent reasoning, we lack verifiable correctness. Can we afford AI-driven decision-making without reliability guarantees? ✅ Agentic AI & Multi-Agent Systems:- The integration of LLMs into autonomous, multi-agent AI systems is a double-edged sword. On one hand, these systems offer adaptive, cooperative intelligence—but on the other, they introduce complexity, opacity, and safety risks. We need governance models that balance autonomy and oversight. ✅ Responsible AI Development & Deployment:- Many organizations still focus on post-deployment fixes rather than AI safety by design. Alignment techniques today (RAG, constitutional AI, human feedback) remain fragile. We must shift toward "failsafe AI"—AI that degrades gracefully rather than unpredictably. ✅ AI Ethics & Governance:- AI risks—whether misinformation, deepfakes, or algorithmic bias—are no longer just theoretical. Geopolitical competition for AI dominance could further sideline ethical considerations. It is time for a convergence of policy, technical safety, and corporate governance models to ensure AI serves societal progress, not just market incentives. 👩💻 The Path Forward: A Call for Multidisciplinary Collaboration:- AI governance cannot be an afterthought. It must be woven into the DNA of AI systems—across research, regulation, and deployment. As someone deeply involved in AI governance and policy, I believe the future lies in co-regulation—where industry, academia, and policymakers collaborate proactively rather than reactively. ✨ How do we get there? 1️⃣ Bridging the gap between AI development and policy-making. 2️⃣ Building safety-aligned benchmarks for Agentic AI. 3️⃣ Embedding ethical constraints within AI architectures, not just in guidelines. 💡 AI is no longer just a tool—it is a co-pilot in decision-making, shaping economies, politics, and societies. The question is: can we govern it before it governs us? 🔎 Would love to hear your thoughts! What challenges do you see in ensuring AI remains safe, aligned, and trustworthy? #AIResearch #ResponsibleAI #AITrust #AgenticAI #Governance #AAAI2025 #AISafety #AIRegulation #EthicalAI

As AI advances apace, potentially beyond "Slave AI", framing and designing "Friendly AI" may be our best approach. A comprehensive review article on the space uncovers the foundations, pros and cons, applications, and future directions for the space. The paper defines Friendly AI (FAI) as "an initiative to create systems that not only prioritise human safety and well-being but also actively foster mutual respect, understanding, and trust between humans and AI, ensuring alignment with human values and emotional needs in all interactions and decisions." It intends to go beyond existing anthropocentric frameworks. Key insights in the review paper from include: 🔄 Balance Ethical Frameworks and Practical Feasibility. The development of FAI relies on integrating ethical principles like deontology, value alignment, and altruism. While these frameworks provide a moral compass, their operationalization faces challenges due to the evolving nature of human values and cultural diversity. 🌍 Address Global Collaboration Barriers. Developing FAI requires global cooperation, but diverging ethical standards, regulatory priorities, and commercial interests hinder alignment. Establishing international platforms and shared frameworks could harmonize these efforts across nations and industries. 🔍 Enhance Transparency with Explainable AI. Explainable AI (XAI) techniques like LIME and SHAP empower users to understand AI decisions, fostering trust and enabling ethical oversight. This transparency is foundational to FAI’s goal of aligning AI behavior with human expectations. 🔐 Build Trust Through Privacy Preservation. Privacy-preserving methods, such as federated learning and differential privacy, protect user data and ensure ethical compliance. These approaches are critical to maintaining user trust and upholding FAI's values of dignity and respect. ⚖️ Embed Fairness in AI Systems. Fairness techniques mitigate bias by addressing imbalances in data and outputs. Ensuring equitable treatment of diverse groups aligns AI systems with societal values and supports FAI’s commitment to inclusivity. 💡 Leverage Affective Computing for Empathy. Affective Computing (AC) enhances AI’s ability to interpret human emotions, enabling empathetic interactions. AC is pivotal in healthcare, education, and robotics, bridging human-AI communication for more "friendly" systems. 📈 Focus on ANI-AGI Transition Challenges. Advancing AI capabilities in nuanced decision-making, memory, and contextual understanding is crucial for transitioning from narrow AI (ANI) to general AI (AGI) while maintaining alignment with FAI principles. 🤝 Foster Multi-Stakeholder Collaboration. FAI’s realization demands structured collaboration across governments, academia, and industries. Clear guidelines, shared resources, and public inclusion can address diverging goals and accelerate FAI’s adoption globally. Link to paper in comments

⚠️ Can AI Serve Humanity Without Measuring Societal Impact?⚠️ It's almost impossible to miss how #AI is reshaping our industries, driving innovation, and influencing billions of lives. Yet, as we innovate, a critical question looms: ⁉️ How can we ensure AI serves humanity's best interests if we don't measure its societal impact?⁉️ Most AI governance metrics today focus solely on compliance and while vital, the broader question of societal impact (environmental, ethical, and human consequences of AI) remains largely underexplored. Addressing this gap is essential for building human-centric AI systems, a priority highlighted by frameworks like the OECD.AI's AI Principles and UNESCO’s ethical guidelines. ➡️ The Need for a Societal Impact Index (SII) Organizations adopting #ISO42001-based AIMS already align governance with principles of transparency, fairness, and accountability. But societal impact metrics go beyond operational governance, addressing questions like: 🔸Does the AI exacerbate inequality? 🔸How do AI systems affect mental health or well-being? 🔸What are the environmental trade-offs of large-scale AI deployment? To address, I see the need for a Societal Impact Index (SII) to complement existing compliance frameworks. The SII would help measure AI systems' effects on broader societal outcomes, tying these efforts to recognized standards. ➡️Proposed Framework for Societal Impact Metrics Drawing from OECD, ISO42001, and Hubbard’s measurement philosophy, here are key components of an SII: 1️⃣ Ethical Fairness Metrics Grounded in OECD principles of fairness and non-discrimination: 🔹 Demographic Bias Impact: Tracks how AI systems impact diverse groups, focusing on disparities in outcomes. 🔹Equity Indicators: Evaluates whether AI tools distribute benefits equitably across socioeconomic or geographic boundaries. 2️⃣ Environmental Sustainability Metrics Inspired by UNESCO’s call for sustainable AI: 🔹Energy Use Efficiency: Measures energy consumption per model training iteration. 🔹Carbon Footprint Tracking: Calculates emissions related to AI operations, a key concern as models grow in size and complexity. 3️⃣ Public Trust Indicators Aligned with #ISO42005 principles of stakeholder engagement: 🔹Explainability Index: Rates how well AI decisions can be understood by non-experts. 🔹Trust Surveys: Aggregates user feedback to quantify perceptions of transparency, fairness, and reliability. ➡️Building the Societal Impact Index The SII builds on ISO42001’s management system structure while integrating principles from the OECD. Key steps include: ✅ Define Objectives: Identify measurable societal outcomes ✅ Model the Ecosystem: Map the interactions between AI systems and stakeholders ✅ Prioritize Measurement Uncertainty: Focus on areas where societal impacts are poorly understood or quantified. ✅ Select Metrics: Leverage existing ISO guidance to build relevant KPIs. ✅ Iterate and Validate: Test metrics in real-world applications

AI Success Requires More Than Technical Excellence—It Demands Aligned Values As organizations accelerate AI adoption, we often focus on capabilities: speed, accuracy, scalability. But there’s a more fundamental question: Do we trust the provider behind the technology? Trust in AI is built on concrete decisions: How is data protected? What guardrails exist against misuse? Are ethical principles embedded in design? Also, do stated values remain consistent, or shift with market pressures? We’re not just selecting tools—we’re choosing partners whose values will shape outcomes affecting our customers, employees, and stakeholders. These values will get embedded into future infrastructure. Before evaluating features, evaluate values. Before signing contracts, examine track records. The most sophisticated AI system built on misaligned values creates risk, not advantage. The future belongs to organizations that recognize AI deployment as a values decision, not just a technology decision.

Over the past several months I’ve been speaking a lot about values-aligned AI. But we’re still in the early days and many don’t understand what values-aligned AI even means. The most common question: which values? Or rather, whose values? The team and I at Gloo think about it in terms of values that contribute to the holistic wellbeing – or "flourishing" – of every individual. The concept of human flourishing is not new, it can be traced back through the centuries, beginning with Aristotle and continuing today as an area of scientific research. As defined by the recently released Global Flourishing Study, human flourishing means "living in a state in which all aspects of a person’s life are going well" and it encompasses everything from health to relationships to finances to spirituality. But today’s AI models aren’t built with any of this in mind, which is why at Gloo we’re working to change that. One of the most impactful things about The Global Flourishing Study is it gives organizations access to open datasets around areas of wellbeing, based on an extensive body of global research. We’re taking that data and applying it to build our models and technologies. Furthermore, we're using those datasets to create standards and benchmarks to measure Gloo, but also to measure all AI models against this common dataset. The goal…to advance AI in a way that supports human flourishing or, in other words, to create values-aligned AI. Because if the AI doesn’t improve human flourishing, it needs to be fixed. If it degrades the human experience, we cannot and should not use it in our chat, agents or code. Just like if it gave the wrong answer to 2+2 or hallucinated an incorrect statement and presented it as fact. As an engineer – simply put, it’s a bug – and a bug needs to be fixed before release.   For those interested in learning more about the Study and the data, you can visit https://lnkd.in/g9mzZ5BV

🔍 Everyone’s discussing what AI agents are capable of—but few are addressing the potential pitfalls. IBM’s AI Ethics Board has just released a report that shifts the conversation. Instead of just highlighting what AI agents can achieve, it confronts the critical risks they pose. Unlike traditional AI models that generate content, AI agents act—they make decisions, take actions, and influence outcomes. This autonomy makes them powerful but also increases the risks they bring. ---------------------------- 📄 Key risks outlined in the report: 🚨 Opaque decision-making – AI agents often operate as black boxes, making it difficult to understand their reasoning. 👁️ Reduced human oversight – Their autonomy can limit real-time monitoring and intervention. 🎯 Misaligned goals – AI agents may confidently act in ways that deviate from human intentions or ethical values. ⚠️ Error propagation – Mistakes in one step can create a domino effect, leading to cascading failures. 🔍 Misinformation risks – Agents can generate and act upon incorrect or misleading data. 🔓 Security concerns – Vulnerabilities like prompt injection can be exploited for harmful purposes. ⚖️ Bias amplification – Without safeguards, AI can reinforce existing prejudices on a larger scale. 🧠 Lack of moral reasoning – Agents struggle with complex ethical decisions and context-based judgment. 🌍 Broader societal impact – Issues like job displacement, trust erosion, and misuse in sensitive fields must be addressed. ---------------------------- 🛠️ How do we mitigate these risks? ✔️ Keep humans in the loop – AI should support decision-making, not replace it. ✔️ Prioritize transparency – Systems should be built for observability, not just optimized for results. ✔️ Set clear guardrails – Constraints should go beyond prompt engineering to ensure responsible behavior. ✔️ Govern AI responsibly – Ethical considerations like fairness, accountability, and alignment with human intent must be embedded into the system. As AI agents continue evolving, one thing is clear: their challenges aren’t just technical—they're also ethical and regulatory. Responsible AI isn’t just about what AI can do but also about what it should be allowed to do. ---------------------------- Thoughts? Let’s discuss! 💡 Sarveshwaran Rajagopal

The new Responsible AI Impact Report from All Tech Is Human makes one thing unmistakably clear: 2026 must be the year of public-benefit AI. Defined as AI built on public values, public oversight, and public infrastructure. It highlights something those of us in the nonprofit world already know in our bones: civil society is quietly leading the shift from AI “principles” to real, verifiable practice.  These aren’t abstract ideas. They’re the governance infrastructure our sector is already helping build. And the risks are accelerating from synthetic media, AI companions, biometric surveillance, fraud, and widening social impacts. This means that the public interest cannot be an afterthought. For years, my work with nonprofits has centered on responsible AI skills and staying human-centered. As Allison Fine and I wrote in The Smart Nonprofit, values are our sector’s native language. Nonprofits (and Philanthropy) can and should lead ethical, values-aligned and responsible AI. The report makes something else clear: responsible AI skills and capacity building for nonprofits are necessary, but not sufficient. In addition, systems are needed from shared standards, public datasets, open safety tools, and others mentioned in the report.  Nonprofits and communities are closest to the people absorbing AI’s impact. That proximity gives nonprofits both the responsibility and the opportunity to help shape AI toward the public good. I recently saw a meme from an Electronic Arts employee Slack channel about the frenzy of AI adoption. I repurposed it to reflect the spirit of this report — and the spirit of our sector: Who are we? NGOs & Civil Society What do we want? Public-good AI AI to do what? Serve people, not profits When do we want it? Right now 2026 will be defined by who gets to decide what AI is for. Our sector has the values, the proximity, and the public trust to help ensure AI serves the common good. Download the report: https://lnkd.in/eZJZGHve from All Tech Is Human How might your organization take one step toward public-benefit AI in the year ahead?