Duarte Dionísio

Observability of trajectories in organizational offboarding

Sat, 18 Jul 2026 17:28:17 +0100

There is a well-known problem in auditing and compliance that is rarely phrased in these terms: most control mechanisms are designed to monitor points, not trajectories. A point is a verifiable state at a given moment: it is access granted, a document signed, an action recorded in a log. A trajectory is the sequence of points over time, and it is within this sequence, not in any isolated point, that certain patterns of non-compliance truly manifest.

This distinction is not merely academic. It is the difference between a system that can say "this action, at this moment, complied with the policy’ and a system that can say "this set of actions, over this period, and viewed as a whole, constitutes a deviation". Most organisational audit processes are only capable of answering the first question.

In distributed information systems, this problem has a name: the compliance of each node, assessed in isolation, does not guarantee the compliance of the system as a whole. It is possible that each component, when observed at its own control point, is perfectly correct, and yet the aggregate behaviour of the system, when these points are correlated over time, reveals a pattern of risk that no individual audit point was designed to detect. It is not that the audit fails due to negligence. It fails because of its architecture: each auditor sees their own fragment, and the breach is not in any single fragment; it lies in the seam between them.

Organisational offboarding processes are structured exactly in this way. An employee’s departure is not a one-off event, even though HR systems often treat it as such, typically a date, a checklist, a handover meeting. In practice, it is an extended period, often beginning well before the formal notification, during which multiple stakeholders, such as colleagues, team managers, asset managers and systems administrators, make independent decisions regarding resources, access rights and responsibilities previously associated with that individual. Each decision, taken in isolation, may be administratively justifiable. The reallocation of equipment before the official departure date may have a legitimate operational reason. The early assignment of a responsibility may simply reflect planning. None of these actions, viewed in isolation, triggers any compliance alarm.

The problem arises when one looks at the bigger picture. Not at a single action, but at the temporal pattern of various actions converging before the point at which, formally speaking, there would not yet be grounds for taking them. It is precisely the sort of signal that eludes any single audit point, because no single audit point has visibility over the entire trajectory, only over its own segment of it.

The typical institutional response to offboarding is a checklist: revoking access rights, retrieving equipment, transferring documentation, within a set timeframe from the departure date. It is a point-based mechanism which checks, on a specific date X, that a set of conditions has been met. It is useful and necessary, but structurally blind to what happens before date X, which is precisely where the problematic pattern of early reallocation tends to be concentrated.

This has a direct parallel with a central argument regarding data protection: compliance with the GDPR is not achieved simply by ensuring that, at any given moment, the data is formally protected. Compliance is achieved by ensuring that the entire data journey, across all processors and sub-processors involved, adheres to the principles of Article 5 (data minimisation, purpose limitation, integrity) over time, not just at a single point in time. A system may pass all one-off audits and still process personal data in a non-compliant manner, precisely because the breach arises from the distributed flow of data across systems, and no single audit point has sufficient visibility to detect it. The architecture of the problem is the same: replace "organisational resources assigned to a person" with "personal data processed by a system", and the structure of the supervisory failure remains identical.

The technical solution to this type of problem in distributed information systems does not lie in increasing the number of audit points, but in correlating signals over time, across multiple nodes, to detect patterns that no single node reveals. Applied to organisational offboarding, this would mean treating the departure period not as a single date but as a time window to be observed as a whole: recording not only what happens on the formal date, but the sequence of decisions relating to a person’s assets, access rights and responsibilities, starting well before that date, and flagging when that sequence begins to diverge in an anomalous way.

This is not about monitoring people. It is about recognising that the correct unit of analysis, whether in data protection or organisational governance, is rarely the single point. It is the trajectory. And as long as control mechanisms continue to be designed around dates and checklists, they will remain structurally incapable of seeing what only becomes apparent when one looks at the entire sequence.

Perhaps the most useful question an organisation can ask itself is not "have we followed the exit protocol?", but rather: "have we even managed to observe what happens before there is a protocol to follow?".

The cost of not thinking and productivity that goes unmeasured

Thu, 04 Jun 2026 00:57:29 +0100

The public debate on artificial intelligence oscillates, with predictable regularity, between two extremes: enthusiasm for its promise and fear of being replaced. On the one hand, there are leaders and decision-makers who see technology as an inevitable driver of growth for economies that have fallen behind in terms of productivity. On the other, there are economists and analysts who point to actual data – such as productivity growth and indicators that rule out the simple effect of demanding more from people and machines – and conclude that the ‘something spectacular’ heralded by the sector has yet to show up in the figures.

Both sides, however, share an assumption that is rarely questioned: that the value of artificial intelligence is measured by the amount of work it replaces or speeds up. It is this assumption that deserves closer scrutiny.

The personal computer and internet revolution removed barriers to finding, storing and transmitting information. The productivity gains were relatively straightforward because people could simply do more – and do it better – of what they already knew how to do. Generative artificial intelligence automates something of a different nature, namely the very production of cognitive outputs, as it writes, synthesises, reasons and makes recommendations. And it does so with such fluency that it is difficult, at first glance, to distinguish between a competent output and one that is plausible but incorrect.

This is where a tension arises that the debate on productivity tends to obscure. When a system produces a draft, an analysis or a screening decision, someone has to verify that result. This verification requires precisely the specialist knowledge that the tool is supposed to make redundant. The time saved in the development phase is partly, and sometimes entirely, taken up by certification. Not because the technology is poor, but because responsibility for the outcome does not disappear when tasks are delegated. It simply shifts.

This shift has a consequence that is rarely factored into corporate adoption models. Organisations that respond to artificial intelligence by hiring fewer junior analysts, fewer early-career lawyers and fewer entry-level technicians are simultaneously eroding the knowledge base needed to oversee the systems they have adopted. A máquina produz resultados em volume infinito. A capacidade humana de os validar é, por definição, finita e depende de uma competência que se constrói com tempo e exposição. Quando essa cadeia de formação se interrompe, o erro oculto cresce sem que a organização tenha desenvolvido os instrumentos para o detetar.

The debate on the taxation of artificial intelligence, the so-called ‘robot tax’, and proposals for a tax on windfall profits made by companies in the sector indirectly reflects this tension. The strongest arguments against specific taxes point to the risk of diverting investment and penalising innovation. But the deeper issue is another: the existing tax system was designed to capture value generated by labour and consumption. As the share of capital in production increases and that of labour decreases, the tax base weakens regardless of technology. Artificial intelligence accelerates this process, but it did not create it. Making it the specific target is merely treating the symptom.

What is missing from this debate is not more regulation or more taxation as ends in themselves. It is a more fundamental question: what, exactly, is the purpose of artificial intelligence in the process of knowledge creation and decision-making?

The prevailing view is that of the assistant: a tool that executes, summarises and makes proposals, freeing the user to focus on tasks with greater added value. This view makes sense, but it assumes that the user retains the ability to judge what the tool has produced. A data dashboard does not make decisions; it presents structured information that a decision-maker interprets based on their own understanding of the problem. Artificial intelligence, when used as an assistant rather than as an autonomous decision-making system, should function in the same way. The problem is that the fluidity of their results creates an illusion of completeness that traditional dashboards have never been able to achieve. A poorly constructed graph immediately arouses suspicion. A well-written text with a flawed conclusion requires the reader to make an active effort to verify it – an effort that the very fluidity of the text tends to inhibit.

There is a tension here that is neither technical nor economic. It is epistemological. Artificial intelligence, when used as a substitute for cognitive effort rather than as a support for it, does not strengthen the user’s capacity; it may actually weaken it. Not because the tool is bad, but because knowledge is built through the process of reaching a conclusion, not simply by having access to it. A worker who systematically delegates diagnostic reasoning to an automated system is not becoming more productive in the sense that matters; rather, they are gradually losing the ability to question the system when it fails.

The public debate on artificial intelligence therefore needs to make a distinction that is not yet sufficiently clear: between adoption that enhances human capability and adoption that quietly replaces it. The former requires investment in training, critical thinking and verification mechanisms. The latter yields short-term gains and long-term vulnerabilities that only become apparent once an error has already had consequences.

Productivity that goes unmeasured is not the productivity that artificial intelligence has not yet generated. It is the productivity that is lost when the effort to understand is replaced by the comfort of delegating.

References and inspiration

Negócios n.º5747 jornaldenegocios.pt

Contact

Tue, 02 Jun 2026 00:14:29 +0100

Page not found

Sun, 31 May 2026 21:50:53 +0100

About

Sat, 30 May 2026 21:57:16 +0100

...me

My academic trajectory sits at the intersection of technology and regulation. It spans an bachelor's degree in Systems and Information Technology from Universidade Atlântica, a Master's in Business Information Systems jointly offered by Universidade Aberta and Instituto Superior Técnico, and ongoing doctoral research in Web Sciences and Technologies at Universidade de Trás-os-Montes e Alto Douro. This research focuses on distributed semantic observability for regulatory compliance monitoring of distributed web infrastructures.

My professional trajectory began in technical roles within communication and marketing, then moved into project coordination for digital platform development. Responsibility for information security and data protection grew steadily from there, eventually positioning me at the junction of infrastructure governance and regulatory compliance.

...the website

This website was created to share ideas, reflections, and approaches to communication systems and technologies, with a particular emphasis on enterprise decision-support systems. Here, I explore how well-structured information can become an essential strategic asset for organizations.

I aim to contribute to understanding the dynamics between technology, data, and decision support, fostering a critical and up-to-date perspective on the role of information systems within organizational contexts.

Beyond serving as a repository for my academic journey, this website also functions as a platform for sharing my as-yet-unpublished research, relevant experiences, and projects in the field of information technologies.

This is an ongoing personal project initiated in 2020. It began using ConcreteCMS, a robust system I also used professionally over many years, and later transitioned to Bludit, chosen for its operational simplicity, which suits the needs of a personal project.