Atka Legal Consultancy Limited’s Post

Very interesting talks at AI Fringe today. Points that I took away were: 🏔 AI Action Summit 10 Feb 25 in France. Henri Verdier (the French AI Action Summit representative) gave a sneak preview as to the agenda which, besides AI safety, will focus on international governance, the future of work and AI for public interests. France will also be launching an AI safety institute. ✅ FM testing and evaluation isn't easy. Testing techniques are not mature and many risks will only become apparent post-deployment so harm reduction must be an on-going process. ⚖️ A gap between upstream and downstream? I asked the panel how FM developers should balance building safeguards into models to restrict harmful uses against maintaining commercial viability and flexibility for downstream users. The answer is this isn't easy either. Out-of-the-box safety measures cannot completely anticipate the variety of eventual use cases. 🍃 Sustainability. Possibly an area where interests are already aligned. Most of FM developers' opex is currently spent on compute so they are already incentivised to make models smaller and more efficient. 🏭 Impact on industry. The best economic data suggests that AI will lead to a net job increase, however the composition of jobs will change. New lines of business will open up, so companies must identify and take advantage of these. #aifringe #aisafety #aisummit #artificialintelligence

The UK 🇬🇧 and the US 🇺🇸 have teamed up in a groundbreaking agreement to ensure the safe development of advanced AI technology! Both nations have pledged to develop robust safety evaluation methods for AI tools and systems, marking a historic collaboration. AI safety is the defining challenge of our generation, and by working together, we can unlock its potential while addressing risks head-on. The agreement builds upon commitments made at the AI Safety Summit, establishing AI Safety Institutes to evaluate AI systems rigorously. This partnership empowers governments to understand and mitigate AI risks to national security and society. What are your thoughts about this collaboration? Comment 👇🏾 #AISafety #AI #Innovation #Collaboration #Technology #Partnership #UKUS #FutureTech #TechNews #AIResearch

U.S. and U.K. Forge AI Safety Partnership Amid Global Technological Concerns. The United States and Britain have officially joined forces through a new science of AI safety partnership to address the challenges and potentials of artificial intelligence (AI). This strategic alliance, cemented by Commerce Secretary Gina Raimondo and British Technology Secretary Michelle Donelan in Washington, aims to pioneer advanced AI model testing protocols. Stemming from commitments made at the AI Safety Summit in Bletchley Park last November, this collaboration underscores a mutual dedication to mitigating national security risks and societal concerns posed by next-generation AI technologies. As global leaders in establishing government-led AI safety institutes, both nations are set to undertake joint testing exercises and explore international cooperatives to ensure a safe transition into the AI-dominated future. This partnership heralds a new era of technological governance, where the immense capabilities of AI are harnessed responsibly to tackle some of the world's most pressing challenges while firmly grasping the reins of associated risks.  #AISafetyPartnership #USUKTechCollab #ArtificialIntelligence #GlobalTechLeaders #AISecurity

"Facilitating the development of a federated, privacy-preserving national #AI outcomes registry. “That is a very important part of how we are starting to think about #operationalizing the principles of value assessment or #bias #mitigation or #drift,” Dr. Rhew said. 2024 HIMSS Conference unveiled "TRAIN" - Trustworthy & Responsible AI Network (TRAIN), a "consortium of healthcare leaders whose aim is to operationalize responsible AI principles and guidelines, such as those that improve the quality, safety and trustworthiness of AI in healthcare settings." Reference: Becker's Healthcare White Paper, "Responsible innovation: How to elevate clinicians + improve health equity with the ethical application of AI", GovCIO HIMSS 24 Federal Takeaways #responsibleai #aiforall #CIO #HIMSS24 #techequity #responibledesign

🔴 Live Now: Join us for an insightful round table discussion on "Leveraging AI for Legislative Impact Assessment," organized by NOVA IMS, as part of the final results presentation for the project "Artificial Intelligence for Impact Assessment (AI4IA)." In the AI4IA project, MagIC researchers have developed a new approach to impact assessments of national and transposed EU legislation using advanced technologies, particularly Artificial Intelligence (AI). This methodology facilitates: 📊 Faster Comparison between existing and new legislation ⚖️ Increased Legal Certainty regarding legislative proposals 📉 Reduction of Administrative Burdens for citizens and businesses This discussion will explore how AI is integrated into the legislative process to enhance efficiency and clarify legal frameworks. This event offers a valuable opportunity to understand the future implications of AI in legislative assessments and to engage with leading experts in the field. 🔗 Watch the round table discussion live here: https://lnkd.in/dCK6kbrv #AI4IA #ArtificialIntelligence #LegislativeImpact #NOVAIMS #MagICResearch #Research #ImpactAssessment #AI #LegalTech

The Western Cape Government's Department of Infrastructure takes a significant step with the release of its position paper, "Guardrails for Artificial Intelligence in the Department of Infrastructure." This document highlights the need for an AI policy framework covering various sectors like transport, housing, education, health, and digital infrastructure, setting the stage for broader provincial AI policies. Ensuring societal benefits, the policy must address ethical implications, governance, and expert advisory support due to the socio-technical nature of AI. The EMORA Group's AI advisory stands ready to support the government in implementing these crucial policies. With expertise in ethical AI practices and strategic guidance, EMORA Group can aid in the responsible and effective deployment of AI technologies in the public sector, enhancing service delivery and operational efficiency. Congratulations to Tim Parle and the entire team for this forward-thinking initiative. This marks a significant stride in leveraging AI's potential while safeguarding societal interests. #ethicalai #aipolicy #responsibleai #innovation #publicsector

Yesterday, the European Commission announced over a hundred companies as the first signatories of the EU Artificial Intelligence (AI) Pact. Among the signatories are multinational corporations and EU small and medium enterprises from various sectors, including #aeronautics, #IT, #automotive, and #healthcare. The #AIACT came into force on August 1, 2024. However, some provisions will only become applicable at the end of the transitional period. Therefore, the European Commission is promoting the AI Pact, encouraging the industry’s voluntary commitment to anticipate the AI Act and begin implementing its requirements ahead of the legal deadline. The signatory companies commit to at least three core actions: 📌 Developing an AI governance strategy to foster the adoption of AI within the organization and work towards future compliance with the AI Act. 📌 Mapping high-risk AI systems: identifying AI systems likely to be categorized as high-risk under the AI Act. 📌 Promoting AI literacy and awareness among staff, ensuring ethical and responsible AI development. 👉 More Info on the AI Pact : https://lnkd.in/eYgEXMrg? #AIAct #EUAIPact #artificialintelligence #Europeancommission

✨ **U.S. and U.K. Announce Partnership on AI Safety Testing 🤝 The United States and the United Kingdom have inked a Memorandum of Understanding (MoU) to embark on a joint venture aimed at testing advanced AI models. 🌐🔬 This groundbreaking partnership, unveiled by U.S. Commerce Secretary Gina Raimondo and U.K. Technology Secretary Michelle Donelan, sets the stage for a unified effort to harmonize scientific approaches and propel the development of comprehensive evaluation suites for AI models, systems, and agents. 🚀🧠 Under the terms of the agreement, the U.S. and U.K. AI Safety Institutes will join forces, fostering information sharing, close cooperation, and potential personnel exchanges to leverage a combined wealth of expertise. Together, these institutes will chart a common path toward AI safety testing, culminating in joint testing exercises on publicly accessible models. 🤝🔍 This landmark collaboration marks a significant stride forward in global AI research and underscores the importance of international cooperation in shaping the future of artificial intelligence. 🌍💡 #AICollaboration #USUKPartnership #AIResearch #Innovation

EU and U.S. Forge AI Safety and Global Development Pact - Joint AI Safety Commitment: The EU and U.S. affirm increased cooperation on AI safety and governance, focusing on risk management and trustworthy AI technologies. - Dialogue for AI Development: Establishment of a dialogue between the European AI Office and the U.S. AI Safety Institute aims to enhance collaboration and share scientific AI research. - Support for Global AI Applications: The partnership emphasizes AI’s role in addressing global challenges like healthcare, agriculture, and climate change, targeting advancements in developing countries. Read more here https://lnkd.in/dN57g3dA #AI #GlobalDevelopment #AISafety #Technology #EU #US #Healthcare #Agriculture #ClimateChange #Innovation

💥It’s been a brilliant first half of Day 2 at #ATxSummit with #ATxAI! First up was an eye-opening discussion around the proliferation of deepfakes, which threaten digital trust. Panellists agreed on the need to promote innovation while upholding ethical standards and protecting human rights, with technical tools, AI governance, global collaboration, and education cited as key countermeasures. The next session delved into the broader evaluation, testing, and governance of AI systems. Industry experts stressed that while technical solutions like benchmarks, tools, and metrics are necessary, these must be grounded in policy frameworks and account for different use cases and industries. A collaborative multi-stakeholder approach is essential as we collectively strive to harness AI’s full potential. Closing out ATxAI was an engaging discussion on accelerating AI sciences, among leaders of national AI safety and research institutes. The panel emphasised that open sourcing AI models and updating benchmarks to account for cultural contexts can drive progress in AI safety. Such efforts require an ecosystem of regulators, businesses, and individual users to work in harmony so that AI can be made safer for all. Check out our carousel for key takeaways from each panel! 🌟Don’t miss your last chance to join us virtually for more insights from global luminaries: https://lnkd.in/gc2_MCYk -- #ATxSG 2024, jointly organised by IMDA and Informa Tech, comprises two main segments - #ATxSummit (30 – 31 May) and #ATxEnterprise (29 – 31 May). Follow us on social media for the latest updates on ATxSummit. #IMDA #RedefiningTech