Human Intelligence - EP.9 Newsletter

The Standard Nobody Was Keeping

Most organisations have systems to govern AI technology. They can assess model risk, cybersecurity, data quality and regulatory compliance. Yet almost none can prove what happens to the people whose work, roles and futures are changed by that technology.

That is the missing layer in responsible AI transformation: human assurance.

As AI adoption accelerates, critical decisions are increasingly being made informally by middle managers. They are checking AI-generated work, correcting errors, coaching teams and deciding whether an output is safe enough to reach a customer. But these decisions often have no clearly named owner, consistent basis or reviewable audit trail. This is shadow governance: a control system developing beneath the organisational chart while leaders focus on the technology.

The risk is growing. Gartner expects one in five organisations to use AI to flatten their structures, potentially removing more than half of existing middle-management positions. If organisations remove those roles before identifying the judgement and oversight they quietly provide, they do not remove risk. They remove the people who were managing it.

The financial and human consequences are already visible. Allianz announced that up to 1,800 roles could disappear from its travel operations as AI takes over call-centre work. WiseTech announced approximately 2,000 redundancies, close to 30% of its workforce, during an AI transformation that prompted a petition signed by more than half of its Australian engineers. Meanwhile, research from Orgvue found that 55% of leaders who made AI-related redundancies now believe they made the wrong decision. Gartner forecasts that by 2027, half of the companies that attributed job cuts to AI will be rehiring people to perform similar work.

These are not simply technology failures. They are failures of transition: judgement removed before trust was established, oversight reduced before quality was assured, and people treated as a cost rather than recognised as part of the organisation’s control system.

This is why Building Mutuality is developing HI Accreditation, an independent standard for assessing the human side of AI transformation.

HI Accreditation does not promise that roles will never change. It establishes a no-blind-displacement standard: evidence that AI-driven change has been governed, measured and managed in ways that protect human capability, voice and dignity wherever reasonably possible.

The accreditation is built around seven assessment domains, each evaluated through a five-level evidence scale. Mandatory requirements include an AI use-case register, workforce impact assessments before significant deployments, structured employee voice, a redeployment-first protocol, contestable high-impact decisions, managers trained in transition rather than only in technology, and annual Human Intelligence reporting.

Its foundation is the MCCI sequence:

Mutuality → Culture → Capability → Integration

Trust and co-authorship come first. Then organisations create the cultural conditions in which people can question, learn and speak safely. Next, they develop the human capability required for new work. Only then should AI be fully integrated into the organisation.

The future of AI will not be judged only by how quickly organisations adopt it. It will be judged by whether people believed the transition was worthy of their trust.

The capability may be artificial. The trust is human.

Download the full newsletter to explore the recovery roadmap and see how your organisation can move from pressure to possibilityhttps://www.mutuality.com.au/s/HI_Newsletter_Issue9_BM_The-Standard-Nobody-Was-Keeping.pdf

Related Articles