Seeing the People Behind the Work

Talking to tech workers, what strikes me most is how deeply many of them care. They notice subtle biases creeping into datasets. They worry about how algorithms could affect vulnerable communities, critically identifying transparency, accountability, and fairness challenges. There are numerous examples of tech workers stepping up to raise concerns about the potential misuse of AI for surveillance, misinformation, or discriminatory practices. Whistleblowers and internal advocates making good trouble have spurred the creation of ethics boards, transparency reports, and more rigorous testing protocols. The influence of tech workers goes beyond individual actions. Collectively, they can push for systemic changes within companies,  such as more inclusive datasets and better protocols to preempt harm. Coming from those who understand the technical realities of how AI systems are actually made, their collective voice often carries weight thanks to these unparalleled insights.

And yet, caring doesn’t automatically mean tech workers can act upon their concerns. Many have reported raising issues only to be told that deadlines are more important, or that management doesn’t have time to deal with “what-ifs.” Feeling their voice is too small to matter or that retribution is too high a cost, some step back. That tension—the desire to do the right thing versus the constraints of the system—is a recurring theme in AI governance.

The Pressures of Corporate Life

Corporate structures can make ethical action extremely challenging. Companies often prioritize speed, growth, and profit over reflection, especially as they race against their competitors to market new products. Engineers might spot potential harm ahead of deployment, but speaking up can feel risky. Some fear being labeled as difficult or slowing down their careers. In some cases, they also jeopardize their equity – a substantial part of their remuneration package. Others worry that even if they speak out, their concerns will disappear into the hierarchy. These pressures are not about motivation—they are structural. Even the most conscientious tech workers can feel powerless in these environments.

Finding Ways to Make a Difference

Despite these challenges, tech workers find ways to influence outcomes. It can be small actions: documenting design choices carefully, asking questions in meetings, or contributing to internal review processes. Over time, these small efforts can shape the design, testing, and deployment of AI in meaningful ways.

Collaboration amplifies impact. Engineers who work alongside ethicists, legal scholars, and civil society groups often find their actions more impactful. Framing issues in ways that leadership can understand by linking them to technical or business implications, can also turn what feels like an uphill battle into meaningful change.

Aligning Culture And Structure

Corporate culture makes all the difference. Companies that encourage open dialogue, value questioning, and support ethical reflection give tech workers the space to act. In contrast, organizations that prioritize speed, hierarchy, and growth over responsibility make ethical action nearly impossible. External frameworks—standards, regulatory guidance, whistleblower protections—also shape the leeway of tech workers. Done right, these frameworks give tech workers tools and a safety net for raising concerns. Market structure also matters. As the AI industry becomes more consolidated among a handful of tech giants racing to capture market share, safety concerns recede to the background. But when culture and external structures align, tech workers can exercise their conscience fully.

Ethics in Everyday Work

AI isn’t built from memos or principles alone. It’s accruing from thousands of small, everyday decisions: how a dataset is labeled, how a model is tested, how a design choice is explained. Tech workers are at the center of those decisions. Supporting them is essential—not just for companies, but for society at large.

Conclusion

Tech workers are the hidden backbone of AI. They hold insight, responsibility, and a sense of moral obligation that can guide AI toward fairness and accountability. Yet their influence is constrained by corporate pressures, hierarchy, and fear of speaking up. Recognizing both their power and their limits is critical.

If we want AI that truly reflects human values, we need to create conditions where tech workers can act responsibly and be heard. Supportive cultures, clear standards, and whistleblower protection are essential to nurture tech workers as conduits for the industry’s conscience. Tech workers are not just builders—they are moral agents shaping the technology that touches all of our lives. Giving them the tools, voice, and support to act responsibly may be one of the most important steps in AI governance.

Principles are important. They give us a moral compass and help define what society expects from AI. But principles alone cannot guide the work of an engineer or a data scientist who is building (or perhaps more appropriately given the lack of legibility, growing) a model today. They don’t tell you how to measure fairness in a dataset, how to make a recommendation system more transparent, or how to document complex design decisions in a way that others can understand. That is where technical standards come in.

Making Principles Practical

Technical standards take abstract ideas and make them actionable. They provide concrete guidance on things like data labeling, algorithm testing, and documentation. They create a shared understanding of what principles require in practice, setting industry-wide expectations. When a team follows a standard, they can be confident they are taking steps to align with broader principles such as fairness and transparency.

Standards also make it easier to compare and evaluate different AI systems. Regulators, researchers, and civil society can use them to spot outliers and take action. Companies can use them as internal roadmaps. Standards create a level of clarity and consistency that principles alone simply cannot provide.

Collaboration Is Key

Developing useful technical standards is not something any single group can do alone. Engineers bring technical expertise. Legal scholars and ethicists bring insight into societal values and human rights. Civil society voices ensure that standards reflect real-world impacts. All of these perspectives are necessary to make standards meaningful and practical.

International collaboration is equally important. AI development is global, and the companies shaping these technologies operate across borders. Standards created in one country can influence practices worldwide. Organizations like the International Organization for Standardization (ISO) and the IEEE are already working on AI-specific standards. These efforts are promising, but they need to be more inclusive, adaptable, and responsive to the rapid pace of technological change.

A Tool for Accountability

Standards do something even more important: they make accountability possible. When expectations are clear, we can measure outcomes. If a standard defines how to test for bias, we can audit systems and hold organizations accountable. Standards transform abstract commitments into concrete actions that can be monitored, reported on, and improved over time.

It is important to note that standards are not a replacement for laws or ethical reflection. They complement both. Principles define what we value, laws specify what is required, and standards show how to get there in practice. Together, they form a toolkit for building responsible AI.

Turning Ideas Into Action

One of the biggest challenges in AI governance is moving from statements of intent to real-world implementation. Technical standards offer a pathway to get there. They provide engineers and organizations clarity on what actions to take, and they give the public and regulators more confidence that AI systems are being developed responsibly.

Standards are not static. AI is evolving quickly, and standards must evolve with it. That requires ongoing engagement between technologists, policymakers, and affected communities. Standardization is not just a technical exercise—it is a continuous process of governance, reflection, and collaboration.

Why This Matters

Closing the gap between AI principles and practice is not just a technical concern; it is a societal one. Principles are meaningful only if they shape the systems that touch our lives. Technical standards translate those principles into clear instructions for those developing and deploying AI systems on the ground. Standards transform principles and commitments into actionable, measurable, and enforceable practices.

In essence, technical standards turn ideas into action so that the lofty ambitions we set for AI translate into real-world outcomes. Building responsible AI is not about statements on paper. It is about the choices we make every day in design, implementation, and oversight. Standards help individuals and organizations do so in a way that is thoughtful, accountable, and aligned with the values we care about.