AI can slowly shift an organisation’s core principles. How to spot ‘value drift’ early

The steady embrace of artificial intelligence (AI) across the public and private sectors in Australia and New Zealand has led to broad guidance on how to use the new technology safely and transparently, with good governance and human oversight.

So far so sensible. Aligning AI use with existing organizational values ​​makes perfect sense.

But here’s the catch. Most references to “responsible AI” assume that values ​​are like house rules that you can write down once, translate into checklists, and enforce forever.

But generative AI (Gen AI) does not simply follow the rules of the house. It changes the house. The distinguishing feature of GenAI is not that it automates calculations, but that it automates plausible language use.

It writes the summary, rationale, email, policy design and performance feedback. In other words, it produces the texts that organizations use to explain themselves.

When a system can instantly generate confident, professional-sounding reasons, it can quietly change what counts as a “good reason” to do something.

This is where the “value drift” begins – a gradual shift in what feels normal, reasonable or acceptable as people adapt their work to what the technology makes easy and compelling.

Invisible ethical shifts

For example, in the workplace, a manager can use GenAI to craft performance feedback to avoid a difficult conversation. The tone is softer, but the judgment is harder to pinpoint, as is the responsibility.

Or a policy team uses GenAI to provide a balanced justification for a contested decision. The prose is polished, but the real tradeoffs are less visible.

For small businesses, the appeal of GenAI lies in speed and efficiency. A sole proprietorship can use it to respond to customers, write marketing copy or create policies in seconds.

But over time, responsiveness can lead to immediate, AI-generated responses rather than careful human judgment. The meaning of good service is quietly shifting.

None of this requires an ethical violation. The drift happens precisely because the new practice feels useful.

GenAI’s biggest ethical impacts don’t often emerge in the form of a single shocking scandal. They are slower and quieter. A thousand small decisions are made in a different way.

The explanation becomes a little smoother. It becomes a little more difficult to be accountable. And it won’t be long before we’re living with a new normal that we didn’t consciously choose.

If responsible AI use is more than good intentions and neat documentation, we need to stop treating values ​​as fixed goals. We need to pay attention to how values ​​change once AI becomes part of everyday work.

Hidden assumptions

Many of the current guidelines for responsible AI follow a simple model: identify the values ​​you care about, embed them in GenAI systems and processes, and then monitor compliance.

This is necessary but also incomplete. Values ​​are not ‘fixed’ once they are captured in strategy documents or policy templates. They are lived out in practice.

They are reflected in the way people talk, what they notice, what they prioritize, and how they justify trade-offs. When technologies change these routines, values ​​are reshaped.

An emerging line of research in technology and ethics shows that values ​​are not simply applied to technologies from the outside. They are shaped from everyday use, as people adapt their practices to what technologies make easy, visible, or persuasive.

In other words, values ​​and technologies shape each other over time, each influencing how the other develops and is understood.

We’ve seen this before. Social media not only tested our existing ideas about privacy. It changed them gradually. What once felt intrusive or inappropriate now feels normal to many younger users.

The value of privacy did not disappear, but its meaning changed as daily practices changed. Generative AI will likely have similar effects on values ​​such as fairness, responsibility and caring.

In our leadership development research, we explore how we teach emerging leaders to recognize and reflect on these shifts.

The challenge is not only whether leaders apply the right values ​​to AI, but also whether they are able to notice how working with these systems can gradually change what those values ​​mean in practice.

Constant vigilance

The emphasis in New Zealand and Australia on responsible AI guidance is sensible and pragmatic. It includes governance, privacy, transparency, skills and accountability.

But there is still a tendency to assume that once the right principles and processes are in place, accountability is assured.

However, if values ​​change as AI reshapes practice, responsible AI needs a practical upgrade. Principles still matter, but they must be accompanied by routines that ensure ethical judgment remains visible over time.

Organizations should periodically assess AI-mediated decisions in high-stakes areas such as hiring, performance management or customer communications.

They must pay attention not only to technical risks, but also to how the meaning of fairness, responsibility or care can change in practice. And they must make it clear who owns the reasoning behind AI-shaped decisions.

Responsible AI is not about freezing values. It’s about staying accountable when values ​​change.