Human Infrastructure Diagnostic

AI runs on human infrastructure. You can't see it from the inside.

Whether AI gets adopted depends less on licences than on trust, advice networks, emotions, beliefs and values. Mirmara diagnoses this human infrastructure independently, grounded in Organisational Psychology and Social Network Analysis, before you invest in the fix.

Turn the lights on Explore a sample report See what becomes visible
A 30-minute diagnostic conversation.
Hover over the screen to discover employee voices

AI adoption can be messy.

There are adoption asymmetries. People move at a different pace.

It is hard to see how employees really stand on AI in their company:

what is their attitude toward AI technologies, which emotions dominate? And what are the underlying motives behind AI-related decisions?

01 Mirmara brings light to the dark 02 The new vocabulary 03 Why it stays hidden 04 Methodology & report 05 Positioning 06 The next step
01Mirmara brings light to the dark

Every decision you make about AI rests on an assumption about your people. Checking it takes two things that rarely come together: someone they'll tell the truth to, and a method built to measure how people actually respond to AI.

The value lies in finding clarity.

Every organisation has two kinds of infrastructure for AI. The technical one (licences, data, tools) gets checked before every rollout. The human one rarely does: how people feel about AI, what they believe it can do, what they actually do with it, what they consider fair, and the relationships all of this travels through. Yet this is what AI adoption runs on. Mirmara makes this human infrastructure visible.

Whom do employees trust, where does knowledge flow, and where does it stall? What do employees think but never say out loud?

Mirmara delivers a comprehensive diagnostic report that examines these questions in depth and makes them discussable. The report is a structured, data-backed answer to the question: where do our people really stand on AI? And can our human infrastructure carry the AI strategy we have committed to? Among other things, the report supports leadership in recognising and naming AI adoption patterns and prioritising interventions.

Questions Mirmara sheds light on include:
01

Why do individual employees and teams use AI so differently, even though they theoretically have access to the same resources, such as licences and training?

02

Who shapes AI behaviour informally, beneath the org chart?

03

What are the barriers to AI use for my team? Capabilities? Trust? A lack of clarity on strategy, or something else entirely?

04

How should I prioritise interventions?

05

How do I shape messaging around AI strategy if I am not sure about the general sentiment?

Challenges Mirmara helps frame and tackle are:
01

Leaders feel a disconnect between AI adoption expectations from upper management and where their teams are in reality.

02

Leaders want to deliberately develop the culture around AI and foster knowledge exchange. But how?

03

Teams lack a shared language for dealing with AI technology. Where trust (psychological safety) is missing, sensitive topics get handled out of sight.

No matter where your team stands with AI adoption, Mirmara provides orientation.

02The new vocabulary

We see a whole new set of vocabulary emerging that captures employees’ experience with AI at work.

Especially expensive phenomena for organisations are:
/ / /

Infrastructure stays invisible until it comes under strain. These are the places where it shows. The language around AI adoption tends to be framed negatively. But these patterns aren't failures to eliminate. They're signals: evidence of how people are already adapting to AI, and where it can work better.

What can look like workarounds, hesitation or resistance, Mirmara reads as a valuable signal: of emerging practices, informal expertise and pressure points in how work is changing. All of it shows where your human infrastructure is carrying load and where it is starting to give way. Empirical data makes it strategically shapeable.

03Why it stays hidden

Speaking openly about how you use AI can be expensive for employees, psychologically, but also literally.

An independent, outside perspective frees employees from that perceived risk.

Trust between employees and Mirmara is built through a scientific approach. Survey responses are anonymous; network data is confidential and reported only in aggregate.

04Methodology & report

The tools most companies have for gathering data, such as performance reviews, engagement surveys, staff questionnaires and licence activations, measure exactly what they were designed to measure. Mirmara's methodology is built specifically to capture how employees experience and behave around AI, and asks not only “what and how much was used?” but why that choice was made. Most AI readiness assessments check data, tools and skills. Mirmara examines the layer they skip.

Mirmara turns the data into a comprehensive diagnostic report.

See a sample report Interactive example for a fictional company.

Leaders receive a Human Infrastructure Index across five layers.

Mirmara developed the concept of human infrastructure for AI: the relationships, emotions, beliefs, behaviours and values through which AI gets adopted, or quietly doesn't. Human infrastructure is the decisive difference between a workforce that has AI tools and one that knows how to use them best.

Leaders receive a Human Infrastructure Index composed of five layers.

  • Affective · how people feel about AI. Captures dominant emotions, attitudes and sentiment.
  • Cognitive · what employees believe it can and cannot do. Captures mental models and technical understanding of the technology.
  • Behavioural · what employees actually do with it. Captures day-to-day behaviour around AI tools.
  • Relational · the network layer: whom employees trust, whom they turn to for advice, and so on. These are the routes along which knowledge and AI practices travel. Reveals influencers who are central to knowledge exchange, and those at the periphery.
  • Values · measures how far employees see AI use as legitimate and fair. How people think about AI responsibility and ethics, and where policies may clash with personal values.

Mirmara works with a mixed-method approach made up of semi-structured interviews, questionnaires and Social Network Analysis.

The methods are complementary. Each captures an important angle of AI adoption: the questionnaire provides the wealth of data, interviews the depth, and sheds light on the relational layer.

Employee voices stay protected throughout. Survey responses are anonymous, participation in the network analysis is voluntary and confidential, and results are presented only at group level. The process is designed in line with GDPR.

Mirmara diagnoses the human infrastructure your AI strategy depends on.

Sequence matters. Most organisations buy “the solution” first, usually training, coaches or consulting, before they have any clarity on where employees stand on AI. If the barrier to adoption is, say, a moral unease, training will not increase use, because training closes a knowledge gap. Mirmara puts the diagnosis first, so investment can be placed strategically.

The diagnostic report is built from three pillars:

  • Leverage points. Shows where measures bring the greatest return. Internal influencers who carry knowledge exchange, for example, can be deliberately involved in workshops and focus groups. With the data at hand, leaders can develop a feel for the tone their communication needs to strike to reach their team.
  • A gap analysis. Where the team should be, compared with where it stands. It reveals acute pain points, opportunities and quick wins. The gap analysis helps shift resources to where they do the most, and backs those decisions with data.
  • A Human Infrastructure Index across the five layers. Shows where your human infrastructure is load-bearing and where it is under strain. Corporate intelligence at the cultural level: you know where your people stand, and gain a valuable edge over the competition in how you adopt AI.

Every diagnostic is tailor-made. Mirmara adapts to your organisation: data can be gathered for selected teams, within individual teams, or across the whole organisation. Scope and approach we develop together in a first conversation.

05Positioning
Anna Steinkamp, painted portrait

Anna Steinkamp · Founder,
Organisational Psychologist

Mirmara does not sell the intervention; it diagnoses the human infrastructure your AI strategy depends on.

Mirmara works independently. No strategy is implemented, no change programme set up, no training delivered. Mirmara is the diagnostic layer before such measures. You would not roll out AI without checking your technical infrastructure. The human infrastructure deserves the same diagnosis. The mission: to help leaders see the cultural dynamics of AI adoption more clearly, before investments flow.

Who is behind Mirmara

Mirmara was founded by Anna Steinkamp. Anna is an organisational psychologist working at the intersection of culture, AI adoption and network science.

Her work with Ethical Intelligence, a consultancy for the responsible use of AI, included a full diagnostic of internal culture through SNA. That work shaped her interest in how influence, trust, collaboration and informal knowledge flows determine whether organisations can adapt to emerging technologies in practice.

She earned a BSc in Organisational Psychology at the University of Europe (Berlin) and an MSc in Performance Psychology at the University of Edinburgh, where she specialised in psychological safety, employee voice and team dynamics.

06The next step

Want to see the human infrastructure your AI runs on?

A 30-minute conversation. Together we talk through a possible diagnostic approach and what you are seeing in your team or organisation.

Talk to Mirmara