A curated guide to algorithmic literacy for digital mediators and inclusion workers. Browse resources to deepen your understanding of algorithms, and better support the citizens who encounter them every day.
My learning goal is:
My preferred format is:
My preferred language is:
My time investment is:
My priority level is:
AI is not magic — it is a set of statistical pattern-matching techniques that work well in narrow domains but fail in predictable ways. Once you understand the...
Read more →AI is already embedded in Dutch daily life — from municipal algorithms to job-matching tools — and understanding its basics helps you see and name what is...
Read more →AI literacy is not about knowing how to code — it is about knowing how to ask good questions, test assumptions, and facilitate informed debate. That is exactly...
Read more →An AI model is entirely defined by what you show it — change the training data and you change the behaviour completely. There is no 'intelligence' separate from...
Read more →Generative AI is "models all the way down": every model rests on a dataset that is itself the product of earlier models, scraping choices and assumptions. The...
Read more →Every effortless AI interaction sits on top of a planetary-scale system of mining, manufacturing, labour and data extraction. The convenience is real, but so is...
Read more →Digital well-being and privacy are not all-or-nothing — small, concrete actions add up, and knowing your options is the first step.
Read more →Platforms perform a double illusion — convincing consumers that everything is automated, and convincing workers that their tasks are not 'real work'. Both...
Read more →You are not alone — there is a national community of people doing exactly your job who share tools, questions, and experience. Connecting to it multiplies your...
Read more →Your professional training and official tools are all in one place — knowing where they are and using them systematically is itself a skill.
Read more →Data protection law (GDPR) is one of the most powerful tools your clients have — but only if they know it exists and how it applies to their specific...
Read more →Some of the most consequential algorithmic decisions in France are made by public institutions on the most vulnerable citizens — and these systems are...
Read more →Digital transformation is not just a technical process — it is a social and political one with winners and losers, and it is happening to public services your...
Read more →Citizens have concrete, enforceable rights when algorithms make consequential decisions about them — including the right to explanation, the right to human...
Read more →Technology choices in public services are value choices — and those values can and should be debated democratically, not decided by procurement teams.
Read more →AI is a system technology — like electricity or the internet — meaning its effects are not contained but permeate everything. That requires a systemic, societal...
Read more →The Dutch data protection authority has the power to investigate and sanction algorithmic systems — and it is actively using that power. Your clients' rights...
Read more →Transparency is a necessary but not sufficient condition for accountability — the register makes algorithms visible, but visibility only matters if citizens and...
Read more →Dutch citizens have digital rights that are routinely violated or ignored — knowing what those rights are, and what violations look like, makes you a more...
Read more →Dutch digital life has its own specific dynamics — Dutch companies, Dutch government systems, Dutch legal context — and understanding that context makes you...
Read more →Algorithmic management is not a neutral technical tool — it systematically reinforces employer power while hiding it behind the fiction of independence. The...
Read more →You cannot help clients improve their digital skills without first honestly mapping your own gaps. Pix makes that audit structured, progressive, and officially...
Read more →Recommendation algorithms do not just reflect your interests — they actively shape them, creating feedback loops that reinforce existing identities and make it...
Read more →Bias does not have to be intentional to be harmful — an algorithm trained on biased historical data will reproduce and amplify that bias automatically, even...
Read more →Facial recognition systems encode cultural and historical biases into mathematical functions — what counts as 'normal' is always a political and social...
Read more →Digital and media literacy are inseparable — understanding AI is part of understanding how media shapes our information environment.
Read more →The digital divide is not just about access — it is about the power to understand, critique, and shape the digital environment you live in.
Read more →AI does not exist in a vacuum — every sector has its own specific AI dynamics, risks and opportunities. A DIW who understands AI in health contexts is better...
Read more →Generative AI represents a qualitative shift from earlier AI — it produces original-seeming content rather than just classifying inputs — and this changes both...
Read more →Complex academic ideas about AI can be explained clearly in 15 minutes by someone who really understands them — and understanding the academic framing helps you...
Read more →The mathematics behind AI are not beyond non-specialists — with good visualisation and pedagogy, the core ideas become genuinely graspable.
Read more →A chatbot answer is not weightless: behind it sits a long material chain of mining, manufacturing, energy and human labour. Making that chain visible turns "is...
Read more →An LLM is not an isolated tool but a node in a web of ecological and social dependencies. Looking at the "ecologies" around a model reframes it as something...
Read more →AI literacy can start from conviviality rather than expertise: sitting down together to talk through AI lowers the barrier and makes the subject approachable...
Read more →The pictures we use to talk about AI shape how people think about it: robot-and-blue-brain clichés make AI seem magical and autonomous. Honest imagery is part...
Read more →A generative-AI tool is the visible tip of a sprawling territory of resource extraction, labour and infrastructure. Cartographing that territory makes plain...
Read more →Technology choices are design choices — and design choices can always be made differently if we choose different values.
Read more →AI development is a geopolitical race with profound consequences for democracy, sovereignty, and human dignity — not just a consumer technology story.
Read more →"Suspicious behaviour" is not something a camera objectively detects — it is a category defined by underpaid human labellers following instructions. The...
Read more →Algorithms embedded in public services (schools, benefits, justice) encode political choices that are presented as technical neutrality — naming that is the...
Read more →The real impact of a public algorithm is rarely what its designers intended; it only becomes visible when the different people it touches compare what they...
Read more →An institution can decide how to live with generative AI democratically: rather than imposing a top-down policy, it can convene the people affected to...
Read more →The most important questions about an AI system come before deployment, not after: a structured self-diagnostic forces an organisation to name the risks to the...
Read more →Some of the highest-stakes algorithms are in public health, and they encode explicit value trade-offs — between fairness, efficiency and what is practically...
Read more →The internet does not have to work the way Big Tech built it — public institutions can choose different values and different tools.
Read more →Digital rights are political rights — they require organised, collective advocacy to defend, and understanding the European policy landscape helps you explain...
Read more →The people building AI are not a monolith — there is a large, active community inside and around tech working toward more ethical outcomes. Connecting to that...
Read more →There are real, working alternatives to Big Tech tools — choosing them is a political act, not just a technical preference.
Read more →Staying current on tech regulation is part of the job — and independent, ad-free journalism is more reliable than tech-company press releases.
Read more →Automated decision-making is spreading rapidly across European public services — knowing the landscape helps you anticipate what your clients will encounter...
Read more →Algorithms can be investigated — they leave traces, produce patterns, and can be tested. Journalism and civic tech can hold them to account.
Read more →Big ideas about technology become accessible when they are told through human stories — and Tegenlicht does this better than almost any other Dutch source.
Read more →Most AI 'solutions' are solutions looking for problems — and the hype cycle serves the interests of investors, not users.
Read more →Public attitudes toward AI are more nuanced and more worried than tech companies acknowledge — and those concerns are grounded in real harms.
Read more →AI bias is not an abstract technical problem — it causes concrete harm to specific people, disproportionately those already marginalised, and naming that harm...
Read more →Rules about digital life do not have to be handed down by experts: people affected by digital exclusion can author their own demands collectively, starting from...
Read more →Data governance is a series of concrete choices with real consequences — and the trade-offs between convenience, privacy, and security become visible when you...
Read more →Data-protection rights are a tool for power, not just privacy: by requesting and pooling the data platforms hold on them, workers can expose management...
Read more →Designing an AI system always requires encoding moral choices — and those choices are not universal, they vary by culture, context, and values. Pretending AI is...
Read more →Opaque algorithmic management systems can be held to account from the outside: when the workers subject to a system pool their own data, they can audit it...
Read more →The "ground truth" a model learns from is a human judgement, not a fact of nature. When the affected community labels the data, it holds power over what the...
Read more →Algorithmic management does not just change pay — it reshapes identity, solidarity, and political consciousness. Workers who understand the system can contest...
Read more →When platforms keep the link between fares and driver pay opaque, aggregating drivers' own data is the only way to see what the algorithm is actually doing to...
Read more →An algorithmic pay change is never neutral: it redistributes income and reshapes workers' agency and well-being, often without notice or explanation. Workers...
Read more →The ethical questions raised by AI are not new — they connect to longstanding philosophical debates about power, agency, and what we owe each other.
Read more →Social media platforms are not neutral tools — they are engagement- maximisation machines that reshape behaviour at scale, and understanding this mechanism...
Read more →The current AI boom is substantially driven by financial speculation, not by genuine usefulness — and understanding that helps you calibrate your own advice...
Read more →Recommendation is editorial: by deciding what to surface to hundreds of millions of people, an algorithm shapes public attention — and that shaping can be...
Read more →Tech ethics is not abstract — it is a series of concrete choices made by real people, and those choices can be contested and changed. Latitudes makes that...
Read more →Different AI systems make different choices — they are not interchangeable, and comparing them reveals the values and constraints built into each.
Read more →Digital competence is a structured, learnable set of skills — not a vague quality you either have or don't. Knowing the framework helps you map your own...
Read more →Neural networks are not magic — they are differentiable functions that minimise prediction error through many small adjustments. Once you see the visualisation,...
Read more →Recommendation algorithms are optimisation problems — and what they optimise for (engagement, not truth or well-being) is a choice that produces predictable...
Read more →The most dangerous AI risks are not science fiction — they are technical properties of current systems (misalignment, deception, hallucination) that are...
Read more →Neural networks need not be entirely unaccountable: in constrained settings their behaviour can be formally analysed and their reliability argued for — a...
Read more →