European Data Compliance Network EUDCN · Company in Belgium Tech

Overview

European Data Compliance Network (EUDCN) operates as a non-profit organization that brings together professionals working in data protection, information security, compliance, and business sectors. The organization functions as a community hub for knowledge sharing and professional networking within these domains.

Key people

Irina Klokova

In the news

Publications and 📰 News from the European Data Compliance Network EUDCN PartnersLinkedIn · 15 Sept 2026
QuotedFARI - AI for the Common Good Institute@fari-ai-for-the-common-good-instituteLast year, we asked for your feedback on the FARI Brussels Conference. And we listened. 👂 Your comments after the 2025 FARI Brussels Conference are helping us make this year’s edition better - from the programme itself to networking and access to the conversations after the event. Here are a few changes you will notice at the 2026 FARI Brussels Conference: 🔹 More time for meaningful discussion We have reduced the number of panel sessions from four to three, allowing more time after each session to explore the topic in greater depth and answer questions from the audience. 🔹 Easier networking Since networking is an important part of the conference experience, participants’ organisations will be included on their badges this year, making it easier to identify relevant connections and start conversations. 🔹 Revisit the sessions All sessions from the 2025 FARI Brussels Conference are available on our YouTube channel: https://lnkd.in/eDC-CMMm And, most importantly, we are still listening. Preparations for the 2026 FARI Brussels Conference — “Brussels, Capital of Artificial Intelligence?” are underway, and there is still time to make this edition even better. 💬 Is there anything else we could improve about the organisation of the conference? Share your suggestions in the comments. 👉 Don't miss it and register now: https://lnkd.in/e7x3pTNc Geoffrey Aerts - Gregory Lewkowicz - Tom Lenaerts - Carl Maria MÖRCH - Laura Jousset - Sezen Avci - Kenneth Hannaert - Rodolphe Doité - Selma EL EMRANI - Université libre de Bruxelles - Vrije Universiteit Brussel
Under the "AI Governance" 🤖 stream by the European Data Compliance Network EUDCN 📚Publications and 📰 News from the EUDCN Partners #ai #aiact #aigovernance #future #knowledgesharingLinkedIn · 15 Sept 2026
QuotedAI & Partners@ai-&-partners📢 𝐀𝐍𝐍𝐎𝐔𝐍𝐂𝐄𝐌𝐄𝐍𝐓: 𝐍𝐞𝐰 𝐄𝐔 𝐀𝐈 & 𝐏𝐚𝐫𝐭𝐧𝐞𝐫𝐬 𝐆𝐮𝐢𝐝𝐞 — 𝐏𝐫𝐨𝐜𝐮𝐫𝐞𝐦𝐞𝐧𝐭 𝐆𝐮𝐢𝐝𝐞𝐥𝐢𝐧𝐞𝐬 𝐟𝐨𝐫 𝐀𝐈 𝐂𝐲𝐛𝐞𝐫𝐬𝐞𝐜𝐮𝐫𝐢𝐭𝐲 𝐢𝐧 𝐇𝐞𝐚𝐥𝐭𝐡𝐜𝐚𝐫𝐞 AI & Partners has published practical guidance for embedding AI CYBERSECURITY into the procurement processes of hospitals and HEALTHCARE providers, translating the European Commission's EU AI Act, NIS2, MDR/IVDR, GDPR, the EHDS Regulation and the Cyber Resilience Act into a coded set of measures across the whole procurement life cycle. If your organisation is procuring diagnostic AI, an AI-enabled MEDICAL device, a generative clinical documentation tool or even an off-the-shelf administrative CHATBOT, this is written for you. 🔍 𝑭𝒐𝒖𝒓 𝑨𝑰-𝒔𝒑𝒆𝒄𝒊𝒇𝒊𝒄 𝒓𝒊𝒔𝒌𝒔 𝒄𝒐𝒏𝒗𝒆𝒏𝒕𝒊𝒐𝒏𝒂𝒍 𝑰𝑪𝑻 𝒑𝒓𝒐𝒄𝒖𝒓𝒆𝒎𝒆𝒏𝒕 𝒅𝒐𝒆𝒔𝒏'𝒕 𝒄𝒐𝒗𝒆𝒓 1. DATA poisoning and adversarial evasion — manipulated training data or crafted inputs that corrupt a diagnostic model's behaviour, sometimes without being visible to the clinician 2. MODEL inversion, membership inference and weight theft — attacks that reconstruct patient training data or exfiltrate the model itself 3. PROMPT injection and jailbreaking — instructions hidden in a document, email or clinical note that a generative system follows as if they came from the user 4. An insecure AI supply chain, plus model drift and automation BIAS — pre-trained models, third-party datasets and open-source ML libraries whose provenance is hard to verify, and silent performance decay if nobody is watching 📋 𝑻𝒉𝒆 𝒑𝒍𝒂𝒏-𝒔𝒐𝒖𝒓𝒄𝒆-𝒎𝒂𝒏𝒂𝒈𝒆 𝒍𝒊𝒇𝒆 𝒄𝒚𝒄𝒍𝒆, 𝒇𝒐𝒖𝒓 𝒎𝒆𝒂𝒔𝒖𝒓𝒆 𝒈𝒓𝒐𝒖𝒑𝒔 General and strategic (AI-GEN) — 11 measures: AI risk in governance, an AI bill of materials, red-teaming, secure architecture, safe non-AI fallback and AI literacy Plan (AI-PLAN) — 7 measures: AI risk identification, legal classification, an AI supply-chain policy, cybersecurity requirements and contractual clauses written into the tender Source (AI-SRC) — 6 measures: supplier selection, DPIA/algorithmic impact assessment, verification of technical security claims, supply-chain integrity checks Manage (AI-MNG) — 5 measures: vulnerability disclosure, asset inventory, incident notification, continuous drift monitoring, secure decommissioning. ---------------------- 𝐓𝐡𝐞 𝐩𝐫𝐨𝐜𝐮𝐫𝐞𝐦𝐞𝐧𝐭 𝐪𝐮𝐞𝐬𝐭𝐢𝐨𝐧 Does your AI procurement process: A) treat AI like any other IT purchase — no AI-specific risk assessment, no supply-chain checks, security requirements bolted on after go-live if at all B) embed AI cybersecurity from the first RFP — provenance, human oversight, incident notification and data-deletion obligations specified before contract award Comment A or B. 👇 #AIForBusiness #Digital #Technology #Healthcare #HealthFinance
Under the "Privacy Pros Tool Kit" stream by the European Data Compliance Network EUDCN 🙏 Thank you Martin Zwick for this overview #dataprotection #digitalgovernance #privacypros #future #knowledgesharing #eudcnLinkedIn · 14 Sept 2026
QuotedMartin Zwick@martin-zwick-a49258214Top 5 GDPR Fines in August 2026 1️⃣ The Netherlands: Uber B.V. and Uber Technologies Inc.: €824,990,000 The Dutch DPA fined Uber for fully automated decisions that deactivated driver accounts based on suspected fraud or low ratings. The decisions were made without human intervention and prevented affected drivers from earning income through the platform. Violated law: Art. 22 GDPR. 2️⃣ Spain: XFERA MÓVILES S.A.U.: €200,000 The Spanish DPA fined the telecommunications provider after a duplicate SIM card was issued to an unauthorised third party. Despite discrepancies in the identity document, the required verification checks were not completed. Violated law: Art. 6(1) GDPR. 3️⃣ Romania: AMATO BESTSELLER S.R.L.: RON 285,395 (~€54,400) The Romanian DPA imposed four fines for inadequate internal instructions, unauthorised access by current and former employees, transparency failures, excessive processing and the unlawful handling of health data. The company also made automated commercial calls without prior express consent. Violated law: Art. 5(1)(c), Art. 9, Art. 14, Art. 32(4) GDPR and national electronic communications law. 4️⃣ Norway: Lab Pharma AS: NOK 205,000 (~€18,400) The Norwegian DPA fined Lab Pharma for breaching its duty to cooperate. The company threatened authority employees with legal action and repeatedly failed to provide requested documents on time. Violated law: Art. 31 GDPR. 5️⃣ Romania: GEROCOSSEN S.R.L.: RON 26,236.50 (€5,000) The Romanian DPA imposed a fine after a cyberattack exposed identification and contact data. The company had failed to implement appropriate security measures and was ordered to improve access logging and monitoring. Violated law: Art. 32(1)(b), Art. 32(2) GDPR. I continue to publish a monthly overview of the top GDPR fines across Europe. This overview covers fines first officially published by the competent authorities in August 2026. If you are aware of any additional fines, please add them in the comments.
Under the "AI Governance" 🤖 stream by the European Data Compliance Network EUDCN 🙏 Thank you Muneendra B for sharing #ai #aigovernance #future #knowledgesharingLinkedIn · 14 Sept 2026
QuotedMuneendra B@muneendrabDept of Science, Innovation and Technology has quietly published an AI Risk Management Toolkit on 08/09/2026, and it's actually worth a read even if you're not in government. It sets out 9 risk categories for AI (financial, legal, transparency, fairness, accountability, contestability, technical robustness, security, and impact on people/environment) A risk appetite matrix from averse to eager, a proper likelihood/impact scoring model with defined probability ranges, and A long appendix of risk identification questions plus treatment options. I've already found a use for it. The categories map closely onto how I structure AI vendor and tool reviews, so the appendix questions are a good sense check when a vendor's questionnaire answers feel not sufficient. It's also built to sit alongside the Cyber Assessment Framework, which makes it easy to tie back to NIST CSF 2.0 when I'm writing things up. It's written for public sector so needs some adapting for a commercial TPRM context, but the structure holds up and it's free. Link below if you want to have a look yourself. https://lnkd.in/eY5VRd9H AI Risk Management Toolkit: guidance https://www.gov.uk/government/publications/ai-risk-management-toolkit/ai-risk-management-toolkit-guidance
Under the "AI Governance" 🤖 and🚄 "Technology & Society" streams by the European Data Compliance Network EUDCN 📚Publications and 📰 News from the EUDCN Partners #datacompliance #future #knowledgesharing #eudcn #ai #aiact #aigovernance #future #knowledgesharingLinkedIn · 14 Sept 2026
QuotedCentre for Digital Governance@centre-for-digital-governance🎙️ New Podcast: Professor Joanna Bryson on Why "You Don't Have a Relationship With AI — You Have One With Yourself" Joanna Bryson, Professor of Ethics and Technology at the Centre for Digital Governance and one of AI ethics' longest-standing and most influential voices, joined host Andrew Phipps MBA, MSc for a wide-ranging and provocative conversation. Her core argument: we don't actually form relationships with AI, we project familiar human patterns onto a system that lacks the bidirectionality real relationships require. The conversation covers a lot of ground, including: ⚖️ Legal liability for AI-generated harm 🌍 Whether companies like Palantir represent a new form of data colonialism 🤖 Grief bots and the "digiDan" recreation of philosopher Daniel Dennett 🧠 Why Joanna remains an outlier among AI thinkers who believe machines are showing signs of consciousness Rigorous, unflinching, and well worth a listen. 🎧 Listen to the full episode: https://lnkd.in/db2Zw9zk #DigitalGovernance #AIEthics
Under the "AI Governance" 🤖 stream by the European Data Compliance Network EUDCN #ai #aigovernance #future #knowledgesharingLinkedIn · 14 Sept 2026
QuotedStephan Geering@stephangeeeringAre AI agents out of control? I wrote a Substack post (link in comments) about the implications of the recent AI agents incidents: 🚨 This is an early warning shot 📈 There is no plateau in sight 🧭 This is less of a cybersecurity issue and more of an alignment issue 🚧 Alignment is unresolved 🏛️ We need tighter frontier AI regulation 🤖 Are we anthropomorphising too much? 🔗 Who is responsible in the supply chain?
Under the "AI Governance" 🤖 stream by the European Data Compliance Network EUDCN #ai #aigovernance #dataprotection #future #knowledgesharingLinkedIn · 12 Sept 2026
QuotedJodi Daniels@jodihoffmandanielsAI has a data problem. It craves company data to satisfy your request, and without boundaries, it keeps digging. This creates risk. Managing it requires evaluating what data a model or tool can access, its intent when it uses that data, and the economic, ethical, and regulatory implications of that intended use. Cillian Kieran, Founder and CEO of Ethyca, says the fix isn't just governing the AI tool. It's governing the data behind it too, because as he puts it, "you can't have AI governance without data governance." To make that happen, companies need to know what data they have collected, where it came from, whether they have sufficient consent, and what legal basis applies to its use. Once companies understand those elements, they can give AI access to the information it needs for a specific purpose and nothing more. So, how can companies put these guardrails into practice while leveraging the value of their data? Tune into this week’s She Said Privacy/He Said Security podcast where Cillian Kieran, Founder and CEO of Ethyca, joins Justin Daniels and I to chat about: 🔷 The connection between AI governance and data governance 🔷 The benefits and risks of using open-source AI models 🔷 Challenges companies face operationalizing AI governance 🔷 AI’s role in streamlining privacy risk assessments 🔷 The importance of human judgment, critical thinking, and questioning AI outputs 🔷 How companies can balance AI data access needs with data minimization principles 🔷 Cillian’s personal privacy and data management tip 🎧 Listen to the full podcast here: https://lnkd.in/eQtckNF2 ♻️ Share this episode with legal, privacy, security, leadership, and business teams. 🛎️ Subscribe to the She Said Privacy/He Said Security podcast.
Under the "Digital Sovereignty" stream by the European Data Compliance Network EUDCN 🙏 Thank you Nandini Sharma for sharing your thoughts #datacompliance #digitalgovernance #privacypros #future #knowledgesharing #eudcnLinkedIn · 12 Sept 2026
QuotedNandini Sharma@nandinisharmaDigital sovereignty is not simply about where our cloud lives. It is about what happens when a dependency fails. I have been reading the European Commission’s proposal for the Cloud and AI Development Act (CADA), and one question stayed with me: How should we really measure digital sovereignty? Data location matters. Jurisdiction matters. European technological capacity certainly matters. But from a cybersecurity and GRC perspective, I believe there is another dimension that deserves equal attention: Systemic Resilience! If a critical cloud provider, AI service, software component or technology dependency suddenly becomes unavailable, constrained or subject to a different legal environment, can an organisation genuinely continue operating? That leads to some interesting questions. - Do we understand concentration risk across our critical technology dependencies? - Are our exit strategies actually executable, or simply well-written documents? - Can workloads and data genuinely be moved when necessary? - Do we understand the material dependencies hidden within increasingly complex AI and software supply chains? And perhaps one that deserves much more attention as we build infrastructure intended to last for decades: Are we designing today for cryptographic agility and post-quantum readiness tomorrow? For me, digital sovereignty should not mean digital isolation. Europe does not necessarily need to build every component itself. But it does need the ability to understand its dependencies, retain meaningful control, substitute where necessary, and continue operating when circumstances change. Perhaps the real test of digital sovereignty is therefore not: “Where does this technology come from?” but: “If this dependency changes tomorrow, can we still operate on our own terms?” The European Commission’s CADA proposal makes this a particularly timely conversation. European Commission: Cloud and AI Development Act (CADA) proposal⁠ [Link attached in my first comment] #DigitalSovereignty #CyberResilience #CADA #AIGovernance #CloudSecurity #DigitalPolicy #Cybersecurity #PostQuantumCryptography P.S. And yes, the visual is AI-generated. It seemed only fair to let AI have a small part in this conversation too. 😃

YBrusselsLocation

Something wrong or missing? Send an update. Fixed within 24 hours.

Overview

Key people

In the news

Related profiles