Nicola Bena’s Post

Seminar Reminder! Second seminar in the series Tales on Data Science and Big Data, organized by the CINI National Lab on Data Science. Date: May 10, 2024 Time: 11:00am-11:30am CET Link: https://lnkd.in/d7H6QWv9 Title: Towards Responsible Data Science: Insights from EU-funded Research Speaker: Francesca Pratesi Abstract: This seminar explores the ethical foundations shaping research practices, particularly within EU-funded projects. Focusing on Horizon 2020 initiatives such as SoBigData++ and TAILOR, we investigate the application of ethical principles within these projects. These projects not only advance specialized research but also aim to equip domain experts with the requisite knowledge and tools for ethical data and AI management. The presentation provides insights into the objectives and methodologies of SoBigData++ and TAILOR, with a specific focus on their ethical dimensions. Emphasis is placed on the challenges encountered and the solutions developed to foster Trustworthy AI, ensuring that AI systems incorporate safeguards to make them reliable, safe, transparent, and respectful of privacy, while also adhering to the FAIR (Findable, Accessible, Interoperable, and Reusable) and FACT (Fairness, Accuracy, Confidentiality, and Transparency) principles. By examining these ethical considerations within these EU-funded projects, we aim to contribute to the discourse on responsible data science and its implications for future research endeavors. #cini #bigdata #datascience #seminar #sobigdata #fair

Did you already register for the #2024 Leuven #AI #Law and #Ethics Conference (#LAILEC)? We have an impressive line-up of panels, keynotes and speakers! Check out the program below. On day 2 we have a panel dedicated to #Generative #AI, #text and #data #|#mining and the #EU #copyright framework to discuss some of the most pressing issues connected to the development of generative artificial intelligence models, including: copyright of inputs and outputs, reservation of text and data mining uses (e.g., #Art. 4 CDSM #opt-outs, as well as other community and industry initiatives), the #AIAct provisions on documentation and publication of sufficient detailed summary, and the role of data spaces and data legislation in improving access to quality AI training data. Speakers: are Marie-Christine Janssens (Moderator), Alain Strowel, Laurent Le Meur, Tamás Szigeti, PhD and Thomas Margoni. https://lnkd.in/eetsexst

🌐 Exciting insights from the Computers, Privacy and Data Protection (CPDP) Conference on AI governance in Brussels! 🔏 The annual CPDP conference brings together top lawmakers, regulators, lawyers, engineers, legal consultants, policymakers, academics and non-governmental organisations from Europe and around the globe. This year's edition, held in May, focused on AI governance, a crucial and timely topic. In our latest guest blog, imec-SMIT, Vrije Universiteit Brussel researcher Ine van Zeeland identifies 5 key trends in #AI governance resulting from the conference: 1. Citizen control over the use of their data for AI development. 2. Independent algorithmic auditing. 3. Improving AI literacy in organisations deploying AI. 4. Oversight and supervision beyond the big players. 5. Synthetic data (somewhat) improve the protection of personal data. Read the full blog via the link: https://lnkd.in/e6-Q3NyC #privacy #AIgovernance #dataprotection #datagovernance

5 trends in AI governance, summarised by the amazing Ine van Zeeland ❤️ and our friends at Knowledge Centre Data & Society 💪🏻 #citizens #algorithmaudits #AiLiteracy #oversight #sytheticdata

ASTHA, thank you so much for your endorsement. I am glad that you found the book both practical and useful. Plus, I love your key takeaways from the book. It helps me to understand what is useful as I write blogs and (hopefully) new books! Thanks again! Packt Dell Technologies Iowa State University - Ivy College of Business The University of Iowa Tippie College of Business Menlo College Coe College Drake University #DataLiteracy #AILiteracy #AIEthics

Check out this post-event wrap-up report on CDAO Government, published by our friends at CDO Magazine! 👀 🎤 "The event proved to be a melting pot of ideas, innovation, and practical insights, aimed at enhancing the efficiency and effectiveness of government services through advanced analytics and AI." 🔗 https://lnkd.in/d8cPzfsU

After attending the Data Science Festival MayDay event last week, from a #ResponsibleAI perspective, the majority of talks (excluding BSI's talk) were underwhelming, especially given the recent changes in the AI regulatory landscape globally. From the talks I have been to, I noticed the following: 🌊 High potential for #ethicswashing (article in collaboration with colleagues at BSI AI Regulatory Services to follow): In one talk, the speaker stressed that "data ethics is an important topic", and then 5 seconds later moves to the next topic. 😍 Following #AI #hype without understanding their data maturity: In one talk about the application of AI in a particular industry, the speaker highlighted the lack of #data infrastructure: which makes me question "is the application of AI here too premature, given the data infrastructure is not in place (promoting rubbish-in-rubbish-out)?". ❓ Responsible AI was very loosely mentioned only once: One talk (finally) spoke about gender bias in #datasets and the impact of this in #ML (e.g. data recycling [https://lnkd.in/ewcRvduk]). However, the main conclusion was "we need society to change", which leads to my next point. 🙈 Having their heads in the sand: I questioned one speaker about measuring data quality by representativeness (https://lnkd.in/eJgZBNeG), to which he responded "data is data" (which is true!), but took no consideration of the fact that these data could be promoting bias. 🤔 Little call to action: In the majority of the talks attended, there were limited calls to action, which developers and business leaders could immediately act on. So, to set the record straight, and to provide a call to action, stay tuned for a webinar (mid-June), where we will dive deeper into what Responsible AI concretely is and is not and what you as business leaders and developers can do to build an ecosystem of #trust, #collaboration and #innovation. If someone was to ask you "Why should I care about Responsible AI?", how would you respond? Jochem Huijps, Marc van Meel, Kolja Verhage, Sjoerd Kampen, Hendrik Pries, Ravit Dotan, PhD, Laura Miller, MA Ethics, Giovanna Jaramillo Gutierrez Ph.D , FHCA, Inma Pérez Ruiz, Gabriella Obispa, Gabriel Udoh, Jérémie Zarka, Carlos Muñoz Ferrandis, Iris Wuisman, Max Dyrhage

Last month I engaged in Strathmore University short course on Data Protection, it was enlightening and opened my mind more in regards to how we protect our data as a growing nation especially in terms of #DataScience and #AI, it is vital to put in place better laws and guidelines and have a greater understanding in general. Here's to more milestones. #Data #DataScience

🌟 Reflecting on JUST Industrial Data Week 🌟 Last Monday I had the honor of delivering the Opening Keynote on “Reaching Trustworthy GenAI for Industrial Applications” at the JUST Industrial Data Week. The discussions throughout the week were incredibly enriching, with diverse perspectives from artists, academics, and industry professionals coming together to explore the future of data ethics and AI. On my train ride back from the week I had started reading the book *Nexus - A Brief History of Information Networks from the Stone Age to AI* by historian (philosopher 🤔) Yuval Noah Harari and came across this second thought-provoking quote in the prologue: > “Our tendency to summon powers we cannot control stems not from individual psychology but from the unique way our species cooperates in large numbers. The main argument of this book is that humankind gains enormous power by building large networks of cooperation, but the way these networks are built predisposes us to use that power unwisely. Our problem, then, is a network problem.” 🤔 This insight is particularly relevant as we navigate the complexities of the Semantic Web, Web of Data, and Semantic Knowledge Graphs. It underscores the importance of adhering to the JUST principles of data: **Judicious, Unbiased, Safe, and Transparent**. Let's continue to build “networks of cooperation” that are not only powerful but also ethical and responsible. Together, we can shape a future where data and AI serve humanity wisely and justly. #DataEthics #AI #GenAI #JUSTPrinciples #IndustrialData #TotalEnergies #SKG #KGA

In 2023, the organisations were focused on adopting big data and AI, often sidelining data governance. As per the https://lnkd.in/gtjFvfYH article, 2024 is expected to be a reflective year with a renewed focus on the ethics of data collection and use, driven by government regulations and consumer pressure. Here is a question: How can organisations ensure that their boards and senior executives fully understand and address data governance challenges, especially in the context of emergent technologies like AI?