Christina Montgomery’s Post

Last November the UK’s flagship summit on AI Safety took place at Bletchley, where governments and tech companies around the world reached a landmark agreement on AI safety co-operation. Six months on the follow-up virtual AI Summit in Korea is now being held and we've seen a number of interesting developments already. On the opening day, 16 global AI tech companies (spanning North America, Asia, Europe and the Middle East) agreed commitments to develop AI safely, described as an historic world first. Each company has pledged to publish a safety framework on how they will measure risks of their frontier models, which will also set out thresholds at which severe risks, unless adequately mitigated, would be deemed intolerable. The thresholds are due to be released ahead of the AI Action Summit in France in early 2025. Global leaders of 10 countries and the EU have also reached an agreement to launch the first international network of AI Safety Institutes to boost cooperation of AI. Other commitments include taking steps towards forging a common understanding on aspects of AI safety and aligning work on research, standards and testing. Further sessions are taking place on safety, sustainability and resilience. #AI #AISeoulSummit #FrontierAISafetyCommitments #SeoulDeclaration https://lnkd.in/eHgT2fKg

Seoul AI Summit - Pioneering Global Standards for a Safer AI Future The AI Safety Summit in Seoul, occurred May 21-22, 2024, was a pivotal event focused on addressing the risks and regulations associated with artificial intelligence. This summit follows the inaugural AI safety meeting held in Bletchley Park, UK, last year. It aims to build on the discussions initiated there, bringing together leaders from government, technology, and civil society to forge a path toward safer AI development and implementation. The inaugural AI safety summit in the UK in 2023 marked a significant step in global efforts to understand and mitigate the risks posed by advanced AI systems. Hosted at the historic Bletchley Park, the meeting gathered tech luminaries, researchers, and officials, including Tesla CEO Elon Musk and OpenAI CEO Sam Altman. They signed the Bletchley Declaration, committing to collaborative efforts to address AI's potential catastrophic risks. This summit in Seoul is part of a broader international effort, including a recent UN resolution advocating for safe, secure, and trustworthy AI, and high-level talks between the US and China on setting shared AI standards. The Seoul summit was co-hosted by the South Korean and UK governments, reflecting a strong bilateral commitment to AI safety. On the first day, leaders from the participating AI companies provided updates on how they have been fulfilling commitments made at the Bletchley summit. The second day featured an in-person meeting of digital ministers, hosted by South Korean Science Minister Lee Jong-ho and Britain’s Technology Secretary Michelle Donelan. Discussions focused on best practices and concrete action plans to mitigate AI's negative impacts on various sectors, including energy use, the labor market, and the spread of misinformation. The Bletchley summit laid a foundational framework for AI safety, but its outcomes were light on specific regulatory details. An interim expert report released ahead of the Seoul summit identifies a range of AI risks, such as fraud, disinformation, and labour market disruptions. South Korea hopes to leverage the summit to take a leadership role in formulating global AI governance and norms. #AISafety #DigitalTransformation #TechLeadership #AIRegulation #GlobalAI https://lnkd.in/g2gFKdih

I already posted this story but this article has some additional information. The UK and the US have inked a MOU to foster collaboration on the development of advanced testing for AI models. This partnership, sealed on Monday 1 April 2024 by Technology Secretary Michelle Donelan of the UK and US Commerce Secretary Gina Raimondo, follows pledges made at the AI Safety Summit in November last year. The agreement underscores a concerted effort by both nations to align their scientific methodologies and collaboratively enhance the evaluation processes for AI technologies. By agreeing to undertake joint testing exercises and encouraging exchanges of expertise between their AI Safety Institutes, the UK and US aim to spearhead a unified approach to AI safety. This initiative is not just about mitigating risks but also about leveraging AI’s potential to improve lives globally. UK Secretary of State for Science, Innovation, and Technology, Michelle Donelan, emphasised the historic nature of this collaboration, viewing it as an extension of the “enduring special relationship” between the two countries. She highlighted the global imperative of AI safety and the collective responsibility to navigate the technology’s challenges and opportunities responsibly.

It's great to see AI tech companies coming together to discuss the safety of their platforms. 16 AI companies from across the globe have committed to not developing or deploying AI models if there cannot be sufficient mitigation of the risks involved. This is a positive and necessary step forward to protect children from the harmful effects of AI, including the generation of deepfake child sexual abuse material. #onlinesafety #saas #safevillage

The second global AI summit in Seoul just took place with the introduction of Frontier AI Safety Commitments. The question is how to map these commitments into transparent, agreed-upon measurement strategies. Perhaps the most notable outcome of the event was the introduction of the Frontier AI Safety Commitments. “These commitments ensure the world’s leading AI companies will provide transparency and accountability on their plans to develop safe AI,” and “it sets a precedent for global standards on AI safety that will unlock the benefits of this transformative technology,” said Rishi Sunak. Additionally, the summit saw the unveiling of the Seoul Statement of Intent toward International Cooperation on AI Safety Science, which focuses on enhancing global cooperation to mitigate AI risks. This statement proposes a framework for sharing crucial data about AI models, their capabilities, and associated risks. The event culminating in the Seoul Ministerial Statement where 27 nations committed to develop shared risk assessments for AI technologies, recognizing the need to address potential severe risks, such as the misuse of AI in acquiring chemical or biological weapons or its ability to evade human oversight. The Seoul Declaration for safe, innovative, and inclusive AI also emerged from the discussions, setting a broad framework for future international collaboration on AI development and safety. Relatedly, the Seoul AI Business Pledge was endorsed by 14 companies, signaling a robust private sector commitment to responsible AI development and the sharing of benefits. Commitments are important, but some of the hardest work is nailing out the details. The next big challenge is in translating these agreements into substantive actions that will shape the future landscape of AI governance. #aiethics #seoul #genai https://lnkd.in/eX23nrsR

A very busy day for AI news. First the EU Council approves the AI Act text, and then we have an agreement on AI safety from the biggest global players in this space. 🌐 On the opening day of the Seoul AI Summit, 16 leading AI tech companies from across the world, including giants from North America, Asia, Europe, and the Middle East, have pledged to uphold safety commitments in the development of AI technologies. 🤝 Spearheaded by the UK and the Republic of Korea, this initiative has brought together industry leaders to agree on a set of safety outcomes that will guide the responsible creation and deployment of AI models. Companies like Zhipu AI (China) and the Technology Innovation Institute (UAE) are among the signatories of the ‘Frontier AI Safety Commitments’. 🔍 These commitments focus on ensuring that AI is developed with accountable governance structures and public transparency, especially concerning frontier AI safety. The agreement also includes provisions for not developing or deploying AI models if the associated risks cannot be sufficiently mitigated. 🌟 Prime Minister Rishi Sunak and Technology Secretary Michelle Donelan MP have expressed that these commitments are a significant step towards establishing global standards for AI safety, which will help unlock the transformative potential of AI for economic growth while managing its risks. #AISafety #AISeoulSummit #ResponsibleAI #GlobalCollaboration #TechNews Jeanne Dauzier, Paul Allen, Mark O'Conor, Danny Tobey, M.D., J.D., Heidi Waem, Kristof De Vulder, Giulio Coraggio, Dr. Thorsten Ammann, Imran Syed, David Booth, Linzi Penman https://lnkd.in/e5-fv-5J

How Safe is AI? 🔍 For every tech enthusiast thrilled about AI, there’s someone else who’s worried about its risks. And they have a point. So, what are we doing about it? The AI Safety Summit in Seoul, held from May 21-22, 2024, was all about tackling these concerns head-on. Building on last year’s summit at Bletchley Park in the UK, this event brought together global leaders, industry giants, and experts to forge a path forward. Here’s what went down: Safety: Reaffirming commitments and mapping out a solid plan for AI safety. Innovation: Balancing the push for innovation with smart risk management. Inclusivity: Making sure AI benefits everyone, everywhere. With big names like OpenAI, Google, Microsoft, and Anthropic in the mix, the summit wasn’t just talk. They laid out specific plans to protect us from AI’s potential downsides and started work on an AI Safety Treaty to ensure consistent global standards. Want to know more about how these decisions will shape the future of AI? Check out the full article for all the key takeaways and what’s next on the horizon. Read the full article here: https://lnkd.in/dthrWjhC #AISafety #AIInnovation #GlobalCollaboration #FutureOfAI #TechForGood Day 22 of #100DaysOfAI

Two key announcements from the UK's AI Safety Institute this month (so far)... 1) 12 April - The UK and the Republic of Korea announced today, they are pressing forward with plans for the upcoming wave of global discussions on the safe development of Artificial Intelligence technologies, with the AI Seoul Summit scheduled for May 21st and 22nd, which will take place partially virtually - read more here: https://lnkd.in/eaS8V5YJ 2) 1 April - The Secretary of State for Science, Innovation and Technology, The Rt Hon Michelle Donelan MP and US Commerce Secretary Gina Raimondo signed a Memorandum of Understanding (MOU) on Monday 1 April 2024, which will see the US and UK AI Safety Institutes work together to develop tests for advanced AI models. Although the two AI Safety Institute announced their cooperation at the AI Safety Summit in Bletchley Park, this memorandum formalises the agreement - read more here: https://lnkd.in/egTt6Tzx #AIsafety #AIsafetyinstitute #AIsafetysummit

🌐 Advancing AI Responsibly: Join the Discussion! 🚀 As the rapid evolution of #AI reshapes our society and industries, navigating its complexities responsibly and safely has never been more important. ❇ I’m thrilled to moderate the panel ▶ Technical, business, and risk factors organisations should consider to advance AI in a responsible, safe, and compliant manner ◀ at the upcoming Australasian Joint Conference on AI Industry Day, by University of Melbourne and RMIT University. 📅 When: Thursday, 28th November 2024 📍 Where: Melbourne Connect, Parkville, Victoria. Our panel will tackle critical questions, including: 💠Balancing innovation and risk in AI adoption 💠Practical steps for AI safety and compliance 💠Addressing knowledge and skills gaps in AI 💠Building trust to enable responsible AI It’s a privilege to be joined by incredible experts: Liming Zhu, Nicola Smyth and Janice Carey from CSIRO, Amazon Web Services (AWS) and Coles Group.   💡 The Industry Day also features an exceptional lineup of speakers and discussions on: ❇Business Risks and Digital Ethics in AI ❇Ensuring AI Compliance and Assurance ❇Building AI Governance and Workforce Skills Aman Verma, Jeannie Marie Paterson, Toby Murray, Simon Wail, Mike Backeberg, Gregory Hill, Eduard Hovy Greg Cameron, James Bailey This event is a unique opportunity to gain practical strategies, network with thought leaders, and prepare your organization for the challenges and opportunities of AI. 👉Learn more and register now: https://lnkd.in/gMrVWk34 📣 Feel free to share this post or tag someone who shouldn’t miss out! #AISafety #GenerativeAI #DigitalEthics #ResponsibleAI #Innovation #AIAustralia #industryDay #AIforBuisness

Last week's AI Seoul Summit brought together leaders from government, industry and academia to hammer out agreements on the responsible development of AI. Over the course of 24 hours, an impressive six agreements, pacts, pledges and statements released, including: 🔹 The Frontier AI Safety Commitments signed by 16 leading AI companies. 🔹 The Seoul Statement of Intent toward International Cooperation on AI 🔹 Safety Science to enable information sharing between national AI institutes. 🔹 The Seoul Ministerial Statement where 27 nations agreed to develop shared risk thresholds for AI. 🔹 The Seoul AI Business Pledge from 14 companies to promote responsible AI practices. While these agreements are non-binding, they represent an important starting point in getting key stakeholders aligned on a shared vision for addressing AI risks and opportunities. Building common ground and initiating collaboration between sectors is a crucial first step. Voluntary commitments may lack enforcement teeth, but they set the stage for the challenging and necessary work ahead to ensure AI aligns with business and innovation incentives while also benefiting and protecting humanity. We look forward to contributing to this critical effort by providing the practical tools to guide organizations in the  advancement of their responsible innovation journey." ➡ Read a breakdown of the AI Seoul Summit agreements: https://lnkd.in/gmKstMDU 🤔 Interested in becoming a RAI Institute member? Learn more in the comments. #ResponsibleAI #AI #AIDevelopment #AISeoulSummit