Week in Review (10/19/24)

Federal

• Labor's AI mission spans public policy and internal agency use The Department of Labor’s Chief AI Officer, Mangala Kuppa, prioritizes AI safety, workforce training, and ethical technology deployment, aiming to enhance public services while safeguarding workers' well-being and addressing capacity challenges.
• AI helps feds catch $1 billion of fraud The U.S. Treasury Department's AI-driven fraud detection has recovered $1 billion in check fraud in 2024, nearly triple last year's results, while helping prevent over $4 billion in fraud overall.
• AI and worker well-being: Principles and best practices for developers and employers The U.S. Department of Labor's new AI principles guide employers in ethically deploying AI while prioritizing worker safety, rights, and empowerment in the evolving workplace.
• Q&A: GSA’s Chief AI Officer on the future of GenAI GSA’s Chief AI Officer, Zach Whitman, highlights the agency’s generative AI pilots, emphasizing their potential to enhance efficiency and collaboration across government through responsible implementation, training, and infrastructure optimization.
• The rise of the Chief AI Officer One year after President Biden’s executive order, federal Chief AI Officers (CAIOs) are driving AI innovation and risk management in government, with agencies adopting varied approaches to the role.
• VA staffers are piloting two chatbots for internal use The VA’s Office of the Chief Technology Officer is piloting two generative AI chatbots to assist with internal administrative tasks, boosting staff efficiency and familiarizing employees with AI tools.
• HUD aims to tackle AI skills gap with new assessment approach The Department of Housing and Urban Development (HUD) launched a new AI skills competency model, using assessments to accurately identify and address AI knowledge gaps among employees, aiming to enhance workforce efficiency and personalized training.
• Army kicks off generative AI pilot to tackle drudge work, hallucinations The Army’s new initiative aims to use generative AI to automate repetitive tasks while addressing challenges like data security and hallucinations, enhancing efficiency in acquisition activities.
• VA is updating its AI suicide risk model to reach more women The VA is updating its AI-driven REACH VET tool to include risk factors like military sexual trauma and intimate partner violence, aiming to improve suicide prevention efforts for women veterans.
• Congressional leaders negotiating potential lame-duck deal to address AI concerns Congressional leaders are negotiating a potential AI bill during the lame-duck period, addressing bipartisan issues like AI research, workforce training, and challenges posed by misinformation and national security.
• Top 5 potential uses, pitfalls for generative AI in federal government Generative AI in federal government offers advantages in policy analysis, security, and knowledge transfer, but pitfalls like security vulnerabilities, poisoning attacks, and system integration challenges require robust, adaptive solutions.
• Partnership for Public Service plans AI training center for federal employees The Partnership for Public Service, supported by a $10 million Google.org grant, plans to open the Center for Federal AI in 2025, providing AI education and career development for federal employees.

State/Local

• New York governor unveils AI research consortium to advance public good Governor Kathy Hochul unveils Empire AI consortium, uniting seven top institutions to collaborate on AI-driven solutions for climate change, food security, health care, and more, prioritizing public good.
• Balancing act: states seek to manage AI’s risks and rewards State legislatures are navigating AI regulations, focusing on balancing innovation with consumer protection. Laws aim to mitigate risks, address discrimination, and establish standards for safe AI adoption across various sectors.
• PA is leading the way in responsible, ethical use of AI that empowers workers Governor Shapiro celebrated Pennsylvania's AI advancements at the AI Horizons Summit, highlighting partnerships with NVIDIA and state universities to foster innovation, responsible AI use, and workforce empowerment across the Commonwealth.
• State and local governments can leverage LLMs for better document management State and local agencies leverage LLMs for efficient document management and enhanced data insights, enabling better decision-making and public services while emphasizing the importance of robust data governance and human oversight.
• What do state governments’ public AI inventories reveal? Thirteen states, including Vermont and Connecticut, are publicly sharing AI tool inventories, disclosing agency usage and impacts on citizens. These inventories highlight transparency in AI applications like fraud detection and cybersecurity.
• With diligence, AI can improve procurement, report says AI can enhance state procurement through automation and fraud detection, but success depends on strategic planning, ethical standards, and cross-agency collaboration, according to a recent report from NASCIO and NASPO.
• S.F. bill calls for inventory of AI tools city uses San Francisco’s proposed bill mandates an AI inventory and impact assessment across city departments, addressing potential job displacement, bias, security, and privacy concerns, while promoting transparency and stakeholder engagement in AI deployment.
• How should Pennsylvania regulate AI? With no federal AI regulations, Pennsylvania is exploring measures like AI usage bans, disclosure requirements, and incident tracking to safeguard residents while balancing innovation and safety in AI deployment.
• AI solved ‘nearly impossible’ task of finding racial covenants in county’s real property deeds Researchers from Stanford and Princeton developed an AI model that rapidly identified racial covenants in Santa Clara County property deeds, saving millions in costs and years of manual labor while ensuring compliance with California’s AB 1466 law.
• Gov. Shapiro: AI is Pittsburgh’s future At the AI Horizons summit, Gov. Shapiro highlighted Pittsburgh’s role in advancing artificial intelligence, emphasizing partnerships with Nvidia and local universities to drive innovation in medicine and robotics.
• Virginia to establish AI task force Governor Youngkin's AI Task Force will advise Virginia on leveraging AI for government efficiency while ensuring safety and fairness, positioning the state as a leader in AI development and innovation.
• AI in Austin – how it’s being used and how they want to regulate As Texas prepares AI regulations, Austin leads in research and implementation. Lawmakers focus on ensuring AI safety, transparency, and ethical use while exploring its potential in government operations and law enforcement.
• AI ‘revives’ victims in Kansas City, Mo., anti-fentanyl campaign Kansas City’s AI-powered "Unfinished Legacies" campaign features fentanyl victims recreated through AI to raise awareness among youth. The campaign aims to humanize overdose statistics and discourage drug use.

International

• G7 Toolkit for AI in the Public Sector A toolkit offers practical guidance for public sector leaders to adopt safe AI policies, featuring ethical practices, G7 insights, and public sector AI case studies.
• Understanding the first wave of AI safety institutes First-wave AI Safety Institutes in the US, UK, and Japan prioritize research, standards, and cooperation. Though impactful, they face challenges around scope, redundancy, and industry relationships.
• UAE's new international AI policy aims to prevent misuse of technology The UAE's five-point AI policy emphasizes transparency, accountability, and international collaboration to ensure responsible AI development, fostering global cooperation to prevent misuse of the technology.
• What can AI do for stagnant productivity in Latin America and the Caribbean? AI and digital technology could drive productivity gains in Latin America, but effective policy is needed to ensure widespread adoption and support transitions in the job market.
• Around the world in AI regulation – how the UK can become a leader in transparency The Public Law Project recommends statutory AI transparency measures, citing global examples to guide the UK in fostering trust and preventing potential injustices linked to public sector algorithms.
• AI in public services – a political panacea? As AI garners political attention in the UK, its role in public services raises questions. Success depends on problem-focused deployment, cautious governance, and a bold vision for AI’s societal impact.
• UK’s Keir Starmer: Don’t be scared of AI UK Prime Minister Keir Starmer promotes a pro-AI stance, encouraging Britain to embrace AI’s economic and public service benefits while balancing regulatory caution, aiming to fuel growth and energy innovation.
• AI adoption in developing countries AI presents vast opportunities for developing countries, driving economic growth, public service improvements, and quality of life, but requires overcoming infrastructure, skills, and inequality challenges through international cooperation and innovation.
• Government use of AI unregulated and open to risk: audit office The Dutch government is using 433 AI systems, but half have not undergone proper risk assessments, raising concerns about privacy and compliance with upcoming EU regulations.
• MODA develops five-year plan for AI Taiwan's Ministry of Digital Affairs (MODA) has introduced a five-year AI strategy to enhance government services, boost digital inclusion, and support AI start-ups with NT$10 billion in financial assistance.
• Welcome to the age of AI – whether you like it or not, minister says Assistant Minister Patrick Gorman urged Australians to embrace AI’s potential while addressing its risks, emphasizing the need for robust laws to protect privacy, prevent abuse, and foster innovation.
• Three reasons why AI may widen global inequality AI may widen global inequality as wealthier nations benefit more from its advancements, while poorer countries face challenges in managing AI disruptions without proper infrastructure and policy support.
• Generative AI in your daily work  Generative AI tools like ChatGPT can enhance productivity by drafting documents, brainstorming, and summarizing information. However, public sector employees must follow ethical guidelines, avoiding sensitive data and reviewing AI outputs for accuracy and bias.

Other

• How PBS is using AI to improve processes and systems for stations, staff and viewers PBS leverages AI and spatial computing to personalize content, enhance operational efficiency, and ensure data privacy, supporting its mission to broaden media accessibility and improve viewer engagement.
• This AI pioneer thinks AI is dumber than a cat Yann LeCun, Meta’s chief AI scientist, challenges prevailing AI narratives, emphasizing that today’s models are far from achieving true intelligence and critiquing exaggerated expectations of imminent AGI.
• Why corporate integrity is key to shaping the use of AI With AI evolving rapidly, businesses play a critical role in fostering ethical standards. Corporate integrity, including ethical governance frameworks, is essential to responsibly manage AI’s risks and align it with societal values.
• Google turns to nuclear to power AI data centers Google partnered with Kairos Power to harness small nuclear reactors for AI data centers, aiming for cleaner, reliable energy. This marks a growing tech trend to reduce emissions while meeting rising energy demands.
• State of AI Report 2024 The 2024 State of AI Report explores advancements in AI research, industry growth, geopolitical impacts, and safety. Key themes include multimodal foundation models, shrinking model performance gaps, and AI’s burgeoning enterprise value.
• DeepLearning.AI TheBatch From Malaysia’s data center expansion to the FTC’s crackdown on deceptive AI practices, and a new AI report highlighting trends and breakthroughs, 2024 reveals AI’s growing influence across sectors and regions.
• Mistral releases new AI models optimized for laptops and phones French startup Mistral introduces “Les Ministraux,” small generative AI models for edge devices, enabling on-device, privacy-focused applications in translation, analytics, and robotics, with flexible licensing options.
• Amazon to use nuclear power to meet AI energy needs Amazon’s AWS partners with X-Energy and Energy Northwest to deploy Small Modular Reactors, targeting 5 GW of nuclear capacity by 2040, reflecting tech’s shift toward reliable, carbon-free energy for AI.
• Apocalypse later?  The real impact of AI on the 2024 elections. Despite concerns over AI’s potential to disrupt the 2024 U.S. elections, experts find its influence mainly amplifies existing threats like misinformation and deepfakes, increasing public distrust rather than introducing new dangers.
• Regulating under uncertainty: Governance Options for GenAI Governments face the challenge of balancing AI's benefits with its risks. As countries like the U.S., EU, and China enact varied regulatory frameworks, transparency, enforcement, and global cooperation are crucial for effective AI governance.
• Policy alignment on AI transparency A new report by Partnership on AI emphasizes the need for global coordination on AI policy frameworks to avoid fragmentation. Key recommendations focus on aligning documentation requirements to ensure transparency and accountability across borders.
• Why AI is going nuclear Google, Microsoft, and Amazon are investing in nuclear energy to power their AI operations. AI’s growing energy demands make nuclear an attractive, reliable, and carbon-free solution for data centers.
• The AI governance arms race: from summit pageantry to progress? Global AI governance efforts remain fragmented, dominated by high-profile summits and symbolic agreements. Moving from voluntary norms to enforceable commitments is crucial for meaningful, unified regulation to ensure ethical AI development.