Innovate UK Business Connect’s Post

Responsible AI Community of Interest April Meeting - Call for Participants!📣👥 The Department of Defense DoD Chief Digital and Artificial Intelligence Office (CDAO) RAI division has created a Responsible AI Community of Interest (COI) to help facilitate adoption of responsible AI (RAI) practices across the DOD, and to build a community around this work. The purpose of this RAI COI is to provide a voluntary forum for (1) sharing knowledge and best practices related to RAI in the defense context, (2) presenting in-progress RAI projects, (3) exploring technologies that advance RAI work, and (4) delivering self-defined outputs, such as processes or standards. 💡✍️ The CDAO RAI COI April Meeting will be on April 30th, from 2:00 - 3:00 PM ET. In this session, we will host a panel discussion titled "Understanding Cyber and Adversarial Risks to AI Capabilities" This conversation will delve into the cyber and adversarial challenges associated with AI capabilities, including large language models, through a discussion on recent research and developments in this field. This meeting aims to foster understanding of the critical vulnerabilities that exist in AI capabilities, knowledge which is essential for ensuring the responsible implementation of AI systems in the Department of Defense. 🗓️🔍 Membership of these meetings will include only personnel from the United States Department of Defense, United States Service Branches, and their partners in industry and academia. It is also restricted to those responsible for the acquisition, planning, implementation, sustainment, or usage of AI technologies and tools. Monthly meetings will be conducted, adhering to the Chatham House Rules. 📢🌐 Please submit a complete registration form for approval only if you are interested in participating in the meeting and meet the above-listed criteria - https://lnkd.in/edNiAVth #ResponsibleAI #AI #ArtificialIntelligence #DOD #AISecurity

𝑯𝒐𝒘 𝒊𝒔 𝑨𝒓𝒕𝒊𝒇𝒊𝒄𝒊𝒂𝒍 𝑰𝒏𝒕𝒆𝒍𝒍𝒊𝒈𝒆𝒏𝒄𝒆 𝒔𝒉𝒂𝒑𝒊𝒏𝒈 𝒕𝒉𝒆 𝒇𝒖𝒕𝒖𝒓𝒆 𝒐𝒇 𝒐𝒖𝒓 𝒊𝒏𝒔𝒕𝒊𝒕𝒖𝒕𝒊𝒐𝒏𝒔? In recent years, Artificial Intelligence (AI) has been transforming sectors and challenging paradigms worldwide. Governments, businesses, and military organizations are investing in AI to optimize operations and enhance decision-making. The Brazilian Army has recently announced the approval of the Strategic Artificial Intelligence Directive (EB20-D-02.031) for 2024. This directive provides comprehensive guidelines for the implementation and use of AI, aligned with the Army’s Information Policy. Key Objectives and Benefits: • Enhanced Decision-Making: AI integrated into the Army Information System (SINFOEx) enables precise analysis and event anticipation. • Information Security: Strict standards to protect against cyber threats. • Training and Partnerships: Encourages continuous training and partnerships with industry and academia to develop ethical AI. • Operational Efficiency: Process automation and resource optimization. • Interoperability: Promotes information exchange between AI systems. The Army’s institutional approach to AI also highlights its importance for the civilian sectors, demonstrating how technology can be implemented ethically and securely. This paves the way for advancements in healthcare, education, public safety, and infrastructure. The Strategic AI Directive positions the Brazilian Army at the forefront of technological innovation, strengthening our operational capabilities and contributing to the country’s development. #Innovation #ArtificialIntelligence #BrazilianArmy #DigitalTransformation #InformationSecurity #Technology #Defense

Digital Forensics is a fascinating industry as I learned from Magnet Virtual Summit 2024 today. I learned about different logs and dumps which vary by manufacturer, the most popular ways on acquiring the information and the critical role logs and dumps play in digital forensics investigations. I am looking forward to watching the On-demand content later, clarifying a greater insight into the digital forensics industry, the tools and challenges we can expect in the industry in the near future. One of the topics I was most surprised by was the Daemon volumes and Artificial Intelligence's ability to generate child exploitation content, so I inquired about how we could differentiate actual child exploitation content from AI generated child exploitation content. This could from answers I received be found by file headers although these may be spoofed so there must be something else indicative of this. Learning about a popular way of stealing corporate files by airdropping using Bluetooth and Wi-Fi was also mentioned and captured my interest. I will be sure to do more research on these topics after watching more sessions to note other research points I can look at. #MagnetForensics #MVS2024 #digitalforensics

🔍✨ New Funding Opportunity: EPSRC Sandpit - AI Decision Support for National Security and Defence ✨🔍 The Engineering and Physical Sciences Research Council (EPSRC) is launching an exciting five-day programme aimed at developing AI-driven decision support systems for national security and defence applications. Funding Amount: Up to £3 million, supporting three projects at 80% of the full economic cost. Focus: To foster innovative solutions leveraging AI to enhance national security and defence decision-making. Eligibility: Applicants must be based at a UK research organisation and eligible for EPSRC funding. Please note that businesses do not qualify for this programme. This is a fantastic opportunity for researchers to contribute to critical advancements in national security through AI technology. https://lnkd.in/eAfw2EKn #EPSRC #AIDecisionSupport #NationalSecurity #FundingOpportunity #Innovation

🌐 Between the Layers: DHS Releases Security for AI Framework for Critical Infrastructure The Department of Homeland Security (DHS) has introduced the “Roles and Responsibilities Framework for AI in Critical Infrastructure” — a first-of-its-kind guide crafted with input from industry leaders, public sector experts, and civil society. This framework offers essential, actionable guidance for secure AI deployment across vital U.S. infrastructure like energy, water, and digital networks. Recognizing the complex vulnerabilities AI can introduce, the framework provides targeted recommendations for all key players: - Cloud Providers: Secure environments and monitor for threats. - AI Developers: Use Secure by Design principles and test for biases. - Infrastructure Operators: Maintain comprehensive cybersecurity. - Civil Society & Public Sector: Drive standards, research, and cross-sector collaboration for AI safety. The framework’s goal is clear: to ensure AI strengthens critical services while safeguarding security, privacy, and civil liberties. For more information on the “Roles and Responsibilities Framework for Artificial Intelligence in Critical Infrastructure,” please visit: https://lnkd.in/gCbcDqVp This post is part of our Between the Layer series. Tune in weekly as we share industry insight and thought leadership topics on #Security4AI. #AI #AIFramework #AIRegulation #AIPolicy #GenAI #LLM #AIGovernance

As emerging technologies become increasingly relevant, the Ministry of Defence, the US Defence, and the Canadian Department of Defence share plans to collaboratively pursue research surrounding AI, resilient systems and information domain-related technologies. ⚙ Read the full article here ⤵ https://lnkd.in/eE33gcad

🔒 Navigating the Complex Landscape of GAI (Generative AI) Risks with NIST’s New Guide! 🔒 The National Institute of Standards and Technology (NIST) has released a comprehensive guide (NIST AI 600-1) to help organizations identify and manage risks unique to or exacerbated by Generative AI (GAI). This guide provides a clear lens through which organizations can frame and execute risk management efforts. The guide outlines 12 key risks: 1). CBRN Information: Eased access to harmful information related to chemical, biological, radiological, or nuclear (CBRN) weapons. 2). Confabulation: The production of confidently stated but erroneous or false content. 3). Dangerous or Violent Recommendations: Eased production of and access to violent, inciting, or threatening content. 4). Data Privacy: Leakage and unauthorized disclosure of sensitive data. 5). Environmental: Impacts due to high resource utilization in training GAI models. 6). Human-AI Configuration: Potential issues arising from the interaction of humans and AI systems. 7). Information Integrity: The potential for large-scale disinformation campaigns. 8). Information Security: Increased cyber threats due to lowered barriers for offensive cyber capabilities. 9). Intellectual Property: Unauthorized use of copyrighted, trademarked, or licensed content. 10). Obscene, Degrading, and/or Abusive Content: Eased production of and access to obscene, degrading, and/or abusive imagery. 11). Toxicity, Bias, and Homogenization: Difficulty controlling public exposure to toxic or hate speech, disparaging or stereotyping content. 12). Value Chain and Component Integration: Non-transparent integration of upstream third-party components. As the GAI landscape evolves, this guide will be updated to reflect the changing nature of these risks. Stay informed and manage your GAI risks effectively with NIST’s latest guide! #AI #ArtificialIntelligence #NIST #CyberSecurity #Risk #Security #GenerativeAI

🤖🚨AI is now a US national security priority. The White House has published a National Security Memorandum stating that “AI is likely to impact nearly every area relevant to national security.” Attracting technical talent and increasing computing power are now official national security priorities. The Department of State (DoS), Department of Defense (DoD), and Department of Homeland Security (DHS) “should use all available legal authorities to facilitate the rapid entry and employment of individuals with relevant technical expertise who could enhance the United States’ competitiveness in AI and related fields.” It is now official policy that the U.S. must lead globally in the ability to train new foundational models. All government agencies will work toward advancing these capabilities. https://lnkd.in/gfV5RTPY #AI #national #security Taken from: https://lnkd.in/ghzkJ_ZF

Today, national security threats have a broadened scope to include techno-economic competitiveness and other non-traditional military threats. As a result, the current operational environment for the IC is instead defined by the rapid development of tools, such as artificial intelligence (AI), and the private sector’s dominance in driving these technological advancements. Therefore, technological change must now be the major catalyst of transformation in the IC’s focus, methods, and culture.