Organizational Trust Concepts

We worry about the lock on the front door, but the real risk is the forgotten OAuth token quietly holding the kitchen door open. We all have that precious recipe, stained with turmeric and filled with memories, a symbol of trust and legacy. But when we invite guests into our kitchen and leave that recipe card exposed, we risk losing control over something invaluable. In our businesses, that “secret recipe” is our data and customer confidence. We share access for all the right reasons: collaboration, convenience, innovation. But every API key, integration, or token extends trust a little further, and sometimes too far. The biggest risks don’t announce themselves with a dramatic break-in. They unfold quietly, when trust is extended without enough checks, and we lose sight of who (or what) holds the keys. Or, as Lee Barney (CISO) shared over coffee, “Everything is getting popped: it’s bushfire season for cyber practitioners.” That line stuck with me, and right now, our job isn’t just locking the door but spotting the sparks before they become wildfires. So, what’s our move? It’s time to rethink trust. Not as a static, one-off decision, but as something dynamic. Trust that adapts in real time, shifting with context, behaviour, and risk. Here’s how we keep our secret sauce safe, even as we invite others to the table: 🥄  Hand out only the right spoons (and collect them when done) 🔒  Hide the recipe card (and watch for sneaky moves) 🔁  Practise changing the locks (and respond fast) Dynamic trust means knowing who’s in our kitchen, what they’re holding, and how quickly we can adapt if something goes missing. Curious? Our latest article dives deeper into butter chicken as a service, and what it teaches us about dynamic trust. #CyberSecurity #MakeCyberSimple #DynamicTrust #ZeroTrust #Leadership

🌟 𝗖𝗮𝗻 𝗼𝗿𝗴𝗮𝗻𝗶𝘇𝗮𝘁𝗶𝗼𝗻𝘀 𝘁𝗿𝘂𝘀𝘁 𝗮𝗻𝗱 𝗯𝗲 𝘁𝗿𝘂𝘀𝘁𝗲𝗱, 𝗼𝗿 𝗱𝗼𝗲𝘀 𝘁𝗿𝘂𝘀𝘁 𝗼𝗻𝗹𝘆 𝗲𝘅𝗶𝘀𝘁 𝗮𝘁 𝘁𝗵𝗲 𝗶𝗻𝗱𝗶𝘃𝗶𝗱𝘂𝗮𝗹 𝗹𝗲𝘃𝗲𝗹? 🌟 Great to see our Academy of Management Review article now in print! Oliver Schilke and I challenge conventional wisdom by introducing "organizational actorhood"—a concept that explains how trust can shift between individual and organizational levels based on context. With a fresh lens grounded in micro-institutional and entitativity theory, we provide a framework to understand trust's complexity in partnerships, offering insights into its drivers, implications, and accuracy in decision-making. Available here: https://lnkd.in/gRhTQgzm and in open access: https://lnkd.in/gXVN9e8T #Trust #Collaboration #OrganizationalTheory #Partnerships HKU Business School The University of Hong Kong

✨ As we step into a new year, I find myself reflecting on a question that has stayed with me for over a decade: Why is vulnerability so central to trust yet so underresearched and under-enacted in organizations and communities? I first posed this question publicly at FINT 2012 in Milan, calling vulnerability the most central - and most neglected - concept in trust research. Today, with our Journal of Trust Research Special Issue on Trust and Vulnerability now published, that provocation has become a foundation. I’m deeply grateful to my co-editors and critical friends, Guido Möllering and Simon Schafheitle. Editing this Special Issue with you wasn’t just academic collaboration - it was an exercise in intellectual trust: questioning assumptions, stretching concepts, and staying with discomfort rather than resolving it too quickly. Our five Special Issue papers reveal three core insights: 1️⃣ 𝗩𝘂𝗹𝗻𝗲𝗿𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗶𝘀𝗻’𝘁 𝗷𝘂𝘀𝘁 𝗮 𝗿𝗶𝘀𝗸 𝗰𝗮𝗹𝗰𝘂𝗹𝗮𝘁𝗶𝗼𝗻 — 𝗶𝘁’𝘀 𝗹𝗶𝘃𝗲𝗱 𝗲𝘅𝗽𝗲𝗿𝗶𝗲𝗻𝗰𝗲. Most research treats vulnerability as something people choose when they decide to trust. But our field is shifting toward seeing vulnerability as a constant condition of social life, shaped by context, emotion, and identity. 👉 Trust work means creating spaces where uncertainty can be acknowledged, not buried. 2️⃣ 𝗞𝗻𝗼𝘄𝗶𝗻𝗴 𝘄𝗵𝗼𝗺 𝘁𝗼 𝘁𝗿𝘂𝘀𝘁 𝗶𝘀𝗻’𝘁 𝗼𝗻𝗹𝘆 𝗰𝗼𝗴𝗻𝗶𝘁𝗶𝘃𝗲 — 𝗶𝘁’𝘀 𝗲𝗺𝗯𝗼𝗱𝗶𝗲𝗱 𝗮𝗻𝗱 𝗿𝗲𝗹𝗮𝘁𝗶𝗼𝗻𝗮𝗹. Beyond rational assessment, people rely on practice, history, and moral commitments to navigate vulnerability. Trust is negotiated moment to moment in interactions, language, and behavior. 👉 Leaders and collaborators must pay attention to how vulnerability is shown, shared, and regulated in daily work - not just to policies or surveys. 3️⃣ 𝗧𝗿𝘂𝘀𝘁 𝘁𝗵𝗲𝗼𝗿𝘆 𝗺𝘂𝘀𝘁 𝗺𝗮𝗸𝗲 𝗿𝗼𝗼𝗺 𝗳𝗼𝗿 𝗰𝗼𝗺𝗽𝗹𝗲𝘅𝗶𝘁𝘆, 𝗻𝗼𝘁 𝘀𝗶𝗺𝗽𝗹𝗶𝗰𝗶𝘁𝘆. The issue calls for theories that explain how trust and vulnerability evolve over time, across contexts, actors, and emotions - moving from linear models to richer, process-based thinking. 👉 Trust practitioners need dynamic approaches: sense-making, repair, and continuous trust building rather than one-off interventions. What this means for 2025: The core capacity organizations and communities need isn’t reducing vulnerability - it’s learning how to enact it well, together. Not as exposure without protection. But as a shared practice of acknowledging uncertainty and dependence with courage and structure. If you’re curious how vulnerability gives trust its meaning and why this matters for relationships, organizations, and society this might be your read as we begin the new year. 📄 Introduction to the Special Issue: https://lnkd.in/ec_mVC7j Here’s to practicing vulnerability where it counts. 📸 With Simon and Guido in Helsinki

This paper reconceptualizes “trustworthy AI” by centering the user's perspective, arguing that AI trustworthiness is not an objective trait but a subjective and context-dependent evaluation. 1️⃣ The trustworthiness of AI is not an objective property but a perception formed by users based on their personal judgments. 2️⃣ Users assess whether AI is trustworthy depending on the context in which it is used, such as decision stakes or time pressures. 3️⃣ Three major research areas—interpersonal trust, trust in automation, and risk and trust—offer useful insights into understanding AI trustworthiness. 4️⃣ Many current approaches to trustworthy AI mistakenly assume it can be fully engineered by meeting technical standards, ignoring the user's perspective. 5️⃣ AI trustworthiness depends on how it is developed, how it is used, and how it is governed, with each context shaping users’ perceptions. 6️⃣ Policies and governance efforts should focus more on transparent processes than on fixed performance outcomes. 7️⃣ Developers and policymakers should actively involve users throughout AI development to better understand their needs and perceptions. 8️⃣ Cross-disciplinary collaboration between computer science, social science, and domain experts is essential for building truly trustworthy AI. 9️⃣ Recognizing that trust is subjective and context-specific can lead to more effective and widely accepted AI systems. ✍🏻 Christopher Wirz, Julie L. Demuth, Ann Bostrom, Mariana G. Cains, Imme Ebert-Uphoff, David John Gagne II, Andrea Schumacher, Amy McGovern, Deianna Madlambayan. (Re)Conceptualizing trustworthy AI: A foundation for change. Artificial Intelligence. 2025. DOI: 10.1016/j.artint.2025.104309

We’re in the middle of a fundamental shift: from trust by default to verify then trust. With 90% of content projected to be AI-generated by next year, we need new ways to establish what is - and that depends on context. *Start with something visual, an image or video. Here, we need labels. Provenance tells us what something is, and where it came from. Because visual content travels far from its source, fidelity to origin becomes essential. *Now shift to a customer service experience. When you’re deep in the interaction, you’re not asking “Is this AI?” You’re asking “Can my problem get solved quickly?” Here, a label would interrupt the flow. Trust happens through contextual signals: the legitimacy of the channel, logged-in status, and the system’s ability to deliver consistent, competent help. *Meanwhile, authenticity itself is evolving. Reid Hoffman’s voice on Masters of Scale was partially synthetic. But authenticity here doesn’t mean organic vocal cords; it means fidelity to the intent of a highly produced podcast in his voice. *Finally, consider value. When AI delivers an experience or capability we couldn’t have otherwise, that outsize value becomes its own signal of trust. We trust what proves itself useful, aligned, and consistent with its purpose. Trust today isn’t static or singular. It’s layered, contextual, experiential, and anchored in fidelity to source, intent, and value. Read the full piece → “How Trust Works Now”

As AI agents become more autonomous, the question isn’t just what they can do—it’s how they do it responsibly. Context matters. Sharing the right information at the right time is what builds trust, and trust is the foundation of every interaction. The theory of contextual integrity frames privacy as the appropriateness of information flow within a given scenario. Applied to AI, it means this: an agent booking a medical appointment should share your name and relevant history—not your insurance details. An agent scheduling lunch should use your calendar availability—not expose unrelated emails. Today’s large language models often miss this nuance, sometimes exposing sensitive data unintentionally. Research like PrivacyChecker addresses this by evaluating information flows at inference time, cutting leakage rates in complex, real-world workflows. Complementary efforts in reasoning and reinforcement learning embed contextual integrity into the model itself—teaching it to decide not just how to respond, but whether to share information. This isn’t just academic. It’s practical. It’s about designing systems that align with human expectations, scale responsibly, and preserve trust in every interaction. For a deeper look, check out https://lnkd.in/g_X6wtsu

As we step into 2026, context will be both our greatest challenge and our greatest ally. AI can rapidly generate strategies, products, solutions, content, but it’s context and judgement that determine whether they’re the right ones. The focus is shifting from what we create to why, when, and how we deliver it. The challenge: context itself is no longer stable. It’s fluid, evolving, and sometimes simply wrong. AI chats, agents, and humans can all shift context in subtle ways. If we don’t pause to assess it properly, we risk solving the wrong problem, very efficiently. The AI Daily Briefing, this week gave an example of how AI agents pulling from multiple enterprise data sources surfaced conflicting “truths” not because of technology failure, but because of human error embedded upstream. Raising the questions: which data do we trust when we’re forming decision conditions? And how can we deliberately build this decision context before acting? One method towards achieving such context, strongly advocated for by my good colleague Lars Godejord is from the three-diamond approach often used in software development, with the first diamond being the discovery diamond: Architectural intent. It starts by mastering structured decision making to gain strategic context. Gaining true situational awareness before jumping into solutions. Options are then structured and explored, expanding the solution space through opportunity trees, before pruning back to those that genuinely align with strategic intent. Then comes critical thinking, cognitive pre-mortems, to surface risks and unknown unknowns before decisions are locked in. Lars explains this in more detail here: https://lnkd.in/ekfCUZ2a Alongside this, I’m a strong advocate for a four-quadrant view when creating decision conditions, drawing on quantitative data, qualitative insight, market awareness, and internal expertise. Pulling it all together to gain a broad context understanding. We also can’t overlook experimentation as a fast way for getting to decision context. As Helge Tennø recently shared in his workshop, experimentation helps organisations understand cause and effect, revealing what actually happens when a variable is changed, rather than what we assume will happen, and with Vibe Coding hugely speeding up the ability to create multiple prototypes, experimentation becomes an even greater approach to arrive at a tested and trusted decision condition. Personally I believe that success won’t come from choosing between deep research or fast experimentation. It will come from creating conditions where context is trusted enough to support good judgement, whether decisions are informed slowly, or tested quickly. In a time when we can build almost anything for anyone, decision context may well become the real superpower.

When stakeholders say they don't trust research, they're not talking about your sample size.   We surveyed 147 researchers and stakeholders in Q3. 89% rated trust as critical, but they defined it completely differently.   Here's the disconnect:   → Stakeholders define trust as: Context + business relevance. Show me you understand our constraints. Don't give me blue-sky solutions.   → Researchers define trust as: Rigor + sample size. Show me the methodology was sound. Prove the study is credible.   Both matter. But one unlocks strategic influence, and one keeps you stuck validating button colors.   At our exec dinners, I ask researchers: "What does your CEO care about right now?" Most can't answer. Or worse, their answer has nothing to do with the research they're running.   That's the problem. Research isn't failing because it's not rigorous. It's failing because it's not relevant.   So if you're a product leader, start making demands:   Ask what decision feels risky, and design a study around it. Not a 47-slide deck. An answer that lands in the decision window.   Don't just report findings. Connect them to outcomes I defend to leadership. Dashboard confusion isn't an insight. "Dashboard confusion creates 47 support tickets per week, costing us X sales hours."   The researchers who become strategic partners don't wait for their team to articulate everything perfectly. They ask the right questions. They translate findings into your language. They show up when decisions get made.

One of the most overused words in corporate leadership: trust. This word shows up everywhere on values posters, town-hall slides, and strategy decks. And yet, for something mentioned so frequently, it’s surprisingly misunderstood - and underdeveloped - by many leaders. Francis Frei, a Harvard Business School professor who studies the topic, defines trust as a triangle composed of three traits: we trust people who are #real with us, who make #sound decisions, and who show us they #care. Through this lens, trust isn’t something we have or don’t have. Rather, it’s fluid, nuanced and contextual, built through time, presence, and emotional labour. From an emotional-intelligence perspective, trust lives at the intersection of self-awareness and relationship management. Leaders who understand their own emotional patterns - especially under stress - are more likely to communicate with clarity and care. And those leaders who actively invest in relationships - who ask, rather than assume, and listen, rather than defend - are the ones who create the conditions for better decision-making and more connected bonds with colleagues. From a neuroscience perspective, a lack of trust limits a team’s capacity for clear thinking. Trust is connected to the brain’s reward circuitry, where joy and stress are modulated. When people feel safe and trusted, they think more clearly, collaborate more openly, and take smarter risks. When they don’t, they are on alert - prone to a hypervigilance that shrinks their capacity to be #creative, #communicative, and #engaged.