OECD.AI’s Post

The OECD AI Incidents Monitor Is An Important Step For Safe AI. However, observing an event is not enough to make it Evidence. An observation will become Evidence when someone explains: (i) Why it happens; and more importantly (ii) How it happens. Only then an observation can be useful for policymakers to take the Right corrective actions in order to prevent repeating of such event. I know what I'm talking about. Conceptually it is no different than the Evidence in public health https://lnkd.in/b-dbk6n Since 2006 I repeat that all policies and investments have social impacts (SI) which are differential and in the end all SI are impacts on health & wellbeing. Establishing Causality for SI is essential for evaluation of algorithms and models, and facilitates policymakers in designing and assessment of different policy options and getting the proper corrective action. I have developed Know-How for establishing causality for SI, which should be embedded at the core of all AI algorithms and models and I'm happy to cooperate with all interested parties. Eva Maydell (Paunova) Kai Zenner Monika Queisser-Richaud Ansgar Koene Lucia Loyo Rachel Adams Dasha Silovic Ilze Melngailis Dr. Dennis Hänsel Luis Kun Nathalie Borgeaud Kamran Khan Dr. Kalpana Seethepalli Karina Angelieva Dr. Susann Roth Caroline Kende-Robb Marina Yastreboff

Thx Tony Moroney - an interesting topic. As others have mentioned, trusting any technology blindly is a mistake. Non-semantic syntactic statistical machines. I think Zapier provides a nice brief explanation (simplified) - "The problem is that the large language models (LLMs) and large multimodal models (LMMs) that underlie any AI text generating tool or chatbot like ChatGPT don't really know anything. They're designed to predict the best string of text that plausibly follows on from your prompt, whatever that happens to be. If they don't know the actual answer, they can just as easily make up a string of nonsense that will fit the bill." 😮 https://lnkd.in/dFraBWdJ

Unsurprisingly, AI hallucination has recently taken centre stage. Blind reliance on fabricated data has severe consequences. It is especially problematic for novices who lack the technical expertise to verify such data. “To what extent is AI hallucination acceptable?” and “How much room for error are we willing to tolerate?” - Arthur D. Little. #ai #artificialintelligence #generativeai #dataintegrity #datainfrastructure #data #LLMs #disruptivetechnologies #digitaltransformation #aiethics #responsibleai #digitaldisruption Paidi O Reilly Prof. Dr. Ingrid Vasiliu-Feltes Martin Moeller Aaron Lax Richard Turrin Imtiaz Adam Irene Lyakovetsky🎧🎙 Dinis Guarda Antonio Grasso Dr. Martha Boeckenfeld Ian Jones Nicolas Babin Mike Flache Olivier Gomez (𝐎𝐆) Giuliano Liguori Dr. Marcell Vollmer Birgul COTELLI, Ph. D. Enrico Molinari Efi Pylarinou Bob Shami Prof. Dr.Dominique J.E. Delporte - Vermeiren, PhD., Hon.Dr. Franco Ronconi Dr. Khulood Almani🇸🇦 د.خلود المانع Olivier Kenji Mathurin Zvonimir Filjak Eveline Ruehlin Sally Eaves Patrick Maroney Orlando Francisco F. Reis Dr. Debashis Dutta Victor Yaromin Neville Gaunt 💡⚡️ Nafis Alam Per Brogaard Berggren Hope Frank Jean-Baptiste Lefevre Lionel Costes Anthony Rochand Sergio Raguso

#didyouknow The United Nations has established a new advisory body on addressing risks and opportunities in international governance of artificial intelligence. The High-Level Advisory Body will explore ways to link various AI governance initiatives and, by the end of 2023, make preliminary recommendations on how to leverage AI to accelerate the delivery of the Sustainable Development Goals in the following areas: International governance of AI A shared understanding of risks and challenges Key opportunities and enablers Read more about the UN’s new High-Level Advisory Body and what Secretary-General António Guterres had to say at the launch: https://lnkd.in/gp9AuE-Z https://lnkd.in/eQRDQpiN #UNAIAdvisoryBody #AIGovernance #SustainableDevelopmentGoals #AIforSDGs #GlobalAIInitiatives #AIChallenges #AIOpportunities #UNLeadership #AntonioGuterres #GlobalGovernance #disruptXpodcast #insights #digitalrevolution #disurptTechnolgy #techrevoltion #digitalrevolution #virtualtalkseries

The pace of progress in Artificial Intelligence (AI) has been unlike any previous technology: the UK AI market is predicted to grow to over $1 trillion (USD) by 2035 - Department for Science, Innovation and Technology. In March 2023, the UK Government published its AI regulation white paper, setting out initial proposals to develop a pro-innovation regulatory framework for AI. It has now published its response to the consultation, including a range of commitments: 🔰 Over £100 million to support regulators and advance research and innovation on AI, including Hubs in healthcare and chemical discovery 🔰 Key regulators asked to publish plans by the end of April for how they are responding to AI risks and opportunities 🔰 UK government makes case for introducing future targeted, binding requirements for most advanced general-purpose AI systems The Ada Lovelace Institute's comments on the policy proposals are interesting: https://lnkd.in/eA8ffYij #ai #artificialintelligence #generativeai #llms #largelanguagemodels #airegulation #airisks #aipolicy #aigovernance #responsibleai #aiethics #disruptivetechnologies #digital #digitaldisruption #digitaltransformation Paidi O Reilly Prof. Dr. Ingrid Vasiliu-Feltes Imtiaz Adam Aaron Lax Sally Eaves Irene Lyakovetsky🎧🎙 Martin Moeller Richard Turrin Ian Jones Dr. Martha Boeckenfeld Nicolas Babin Patrick Maroney Birgul COTELLI, Ph. D. Phillip J Mostert 🇿🇦 Orlando Francisco F. Reis Dr. Marcell Vollmer Enrico Molinari Dinis Guarda Efi Pylarinou Franco Ronconi Prof.Dr. Dominique J.E. Delporte - Vermeiren Dr. Monika Sonu Dr. Khulood Almani🇸🇦 د.خلود المانع Olivier Kenji Mathurin Avrohom Gottheil Dr. Debashis Dutta Nafis Alam Hanadi El Sayyed Eveline Ruehlin Knut Jägersberg Amna Usman Chaudhry Olivier LABORDE Harold Sinnott Neville Gaunt 💡⚡️ Glen D Gilmore Devesh Khanna Mike Nash (BA HONS) Per Brogaard Berggren

Defining AI incidents and related terminology The first step is to standardise what constitutes AI incidents and hazards. These definitions are crucial for ensuring a shared understanding and point of reference for regulators across jurisdictions so that they can work together to mitigate risks and hazards with the same reference points. This is what is called interoperability. At the same time, the definitions are drafted to provide flexibility for jurisdictions to determine the scope of AI incidents and hazards they wish to address. #oecd #ocde #ia #ai #risk #incident #definition #terminology #hazards #regulators #privacy #standards

This week, The Royal Society has published recommendations to the United Nations High-Level AI Advisory Body, following a multi-stakeholder dialogue hosted last month. TGI's Shahzeb Mahmood and Sabhanaz Rashid Diya participated in the multi-stakeholder session, highlighting the UN's role in building consensus, elevating unique risks and opportunities posted by AI in the #GlobalSouth and what the minimum guardrails are, particularly in resource-constrained LMICs. The recommendations were shared with the UN ahead of the Advisory Body's meetings in Geneva last week. Read the full report below. #artificialintelligence

New Report! Name it to tame it: Defining AI incidents and hazards 👉  https://lnkd.in/eWqiRyAs We need to use common terminology to describe AI systems' problems and failures so that we can observe, document, report, and learn from them. Today, these events are broadly referred to as "AI incidents". The OECD is leading recent developments in global initiatives that aim to standardise how we report and monitor AI incidents and hazards, paving the way for better governance and safer AI implementations. In this blog post, Luis Aranda and Karine Perset discuss the OECD's proposals for definitions linked to AI definitions and hazards. Special thanks to the Co-Chairs of the OECD.AI Expert Group on Incidents Marko Grobelnik, Elham Tabassi, Irina Orssich, and Mark Latonero, PhD Celine Caira Jamie Berryhill Lucia Russo Noah Oder John Leo Tarver ⒿⓁⓉ Rashad Abelson Angélina Gentaz Valéria Silva Bénédicte Rispal Robin Staes-Polet Eunseo Dana Choi  Pablo Gomez Ayerbe Sara Fialho Esposito Nikolas S. Sarah Bérubé Guillermo H. #aiincidents #trustworthyai #aipolicy #oecd

Latest from FE News: New commitment to deepen work on severe AI risks concludes AI Seoul Summit: 27 nations, including the United Kingdom, Republic of Korea, France, United States, United Arab Emirates, as well as the European Union, have signed up to developing proposals… New commitment to deepen work on severe AI risks concludes AI Seoul Summit was published on FE News by FE News Editor #EdTechNewsonEducationTechnology #Workandleadership Read more here:

As the #EU #AI Act strives to establish a harmonised regulatory framework for artificial intelligence across member states, the accuracy of predictions regarding implementation rigour becomes paramount - AI & Partners Here's why: 🔰 The EU AI Act aims to ensure consistency and effectiveness in AI governance across Europe; predictions based on historical regulatory implementations help anticipate challenges and disparities, enabling proactive measures to foster a cohesive hashtag AI landscape within the European Union. 🔰 Without accurate predictions, variations in implementation rigour among EU countries could undermine trust in cross-border AI applications and impede the free flow of AI technologies; predicting alignment with the EU AI Act allows targeted interventions, ensuring a robust and coherent AI ecosystem throughout the European Union. #ai #artificialintelligence #aipolicy #airegulation #aigovernance #aistrategy #airisks #ecosystem #disruptivetechnologies #digital #digitaldisruption #digitaltransformation Paidi O Reilly Prof. Dr. Ingrid Vasiliu-Feltes Imtiaz Adam Aaron Lax Ian Jones Sally Eaves Irene Lyakovetsky🎧🎙Martin Moeller Dr. Monika Sonu Richard Turrin Dinis Guarda Nicolas Babin Birgul COTELLI, Ph. D. Dr. Martha Boeckenfeld Patrick Maroney Phillip J Mostert 🇿🇦 Orlando Francisco F. Reis Dr. Marcell Vollmer Knut Jägersberg Enrico Molinari Efi Pylarinou Franco Ronconi Avrohom Gottheil Olivier Kenji Mathurin Neville Gaunt 💡⚡️Prof.Dr. Dominique J.E. Delporte - Vermeiren Dr. Debashis Dutta Dr. Khulood Almani🇸🇦 د.خلود المانع Nafis Alam Hanadi El Sayyed Mike Flache Amna Usman Chaudhry Olivier LABORDE Devesh Khanna Jean-Baptiste Lefevre Sergio Raguso Mike Nash (BA HONS) Mario Pawlowski 🇵🇱🇺🇦🇺🇸🇪🇺 ARLENE NEWBIGGING GRADY