- AI Week in Review
- Posts
- AI Week in Review - 1/04/26
AI Week in Review - 1/04/26
Public Sector
Chris Kraft
January 04, 2026
The Details
Federal
Andreessen Horowitz’s federal AI legislative roadmap urges Congress to enact a 9‑pillar framework that protects people and safety while fostering competition and innovation for U.S. AI leadership.
In Capstone 2025, the U.S. Air Force and allied partners tested AI-enabled battle management and joint command-and-control tools to improve dynamic targeting, mission planning, and interoperability across services and nations.
The 2026 National Defense Authorization Act (NDAA) accelerates AI adoption across the Pentagon and intelligence community while imposing new governance, risk assessment, and robust cybersecurity requirements for AI systems and related procurement.
The U.S. Department of Defense’s 2025 China Military Power Report highlights Beijing’s rapid progress on large language models (LLMs) and generative AI for military use, narrowing the U.S. lead and supporting PLA capabilities across cyber, decision support, and influence operations.
About 25,000 people have applied to join the Trump administration’s new federal Tech Force initiative aimed at recruiting AI and tech experts for government roles, with ~1,000 slots available.
AI‑related lobbying has surged in Washington, with registered firms earning nearly $92 million in the first three quarters of 2025 as tech giants and upstarts push Congress and the administration on AI policy.
U.S. AI labs and policymakers must strengthen cybersecurity and reporting standards to prevent Chinese actors from exploiting American AI technologies for surveillance, propaganda, or national‑security harm.
The U.S. Army has created a dedicated 49B AI and ML career field for officers, allowing them to transfer into and specialize in integrating AI/ML across operations with advanced training.
State / Local
New York Gov. Kathy Hochul signed the RAISE Act, requiring frontier AI developers to publish safety frameworks, report critical harms, and create state oversight to set a nation‑leading AI transparency standard.
State governments are expected to expand AI adoption in 2026 with focus on ROI and agentic tools, even as federal executive orders complicate regulatory authority and states pursue varied AI strategies.
Florida lawmakers filed SB 482 to establish an “Artificial Intelligence Bill of Rights” granting residents transparency, parental controls, political-ad disclosures, and restrictions on AI use and contracts with foreign-linked firms.
Mississippi’s Department of Education launched a pilot where 43 teachers across 15 districts use generative AI tools like Google Gemini and NotebookLM to enhance lesson planning and engagement.
Alaska’s court system spent over a year developing an AI chatbot for probate help, but the project faced significant delays and unreliable AI “hallucinations,” underscoring challenges in AI deployment.
A new CA law requires large AI developers to disclose how they assess and manage catastrophic risks and provide whistleblower protections.
New Jersey’s AI Innovation Challenge will award funding, mentorship, and support to startups and teams building AI solutions for state problems, backed by the NJ Economic Development Authority and public‑good goals.
International
Stanford HAI’s brief shows China’s open‑weight AI ecosystem—driven by diverse labs and efficient models now competitive globally—raising key policy questions about governance, competition, and international tech reliance.
Japan will invest ¥1 trillion over five years to launch a national AI champion, uniting firms like SoftBank to develop foundational models, secure chips, and reduce reliance on foreign tech.
Punjab’s government launched AI‑driven career guidance labs in 25 public schools to help students assess aptitudes, explore career paths, and make informed future choices at no cost.
South Korea unveiled a national venture strategy to build 50 unicorns, 10,000 AI/deeptech startups, and attract $27 billion annually in venture investment by 2030.
China’s People’s Liberation Army is building an AI‑enabled “smart logistics” system that uses real‑time sensing, predictive planning, autonomous delivery, and civilian integration to streamline supply flow and support troops.
Taiwan’s Legislative Yuan passed the Artificial Intelligence Basic Act, setting governance principles for AI development, designating the National Science and Technology Council as lead authority, and requiring a national AI strategy committee.
Japan’s Cabinet approved its first national basic AI plan, aiming to balance innovation with trustworthy AI, boost domestic AI development and adoption, and strengthen safety, education, and governance.
Türkiye’s 2026 Presidential Annual Program elevates AI to a central role in governance, strategic autonomy, and defense, embedding it across public administration and resilience efforts to fortify middle‑power capacity.
China’s AI tool PANDA can detect pancreatic cancer early from routine CT scans, identifying dozens of cases missed by doctors and offering potential to improve survival despite false positives.
South Korea’s government‑backed national AI teams publicly unveiled early results, showcasing advanced models and competitive strategies to boost domestic AI capability.
England’s NHS is using a new AI‑powered demand‑forecasting tool in about 50 trusts to predict peak A&E demand, help plan staffing and beds, and reduce winter waits and pressure on emergency care.
Everything Else
Google Cloud’s AI Agent Trends 2026 report forecasts that AI agents will drive enterprise transformation by automating workflows, boosting productivity, enabling interoperable multi‑agent systems, and reshaping work with practical, strategic business value.
The paper mHC: Manifold‑Constrained Hyper‑Connections introduces a framework that stabilizes and scales advanced residual connections in neural architectures by projecting them onto a constrained manifold, improving training stability and efficiency.
AI is reshaping U.S. policing—helping generate reports from body‑cam footage, guide patrols, and analyze surveillance and data center video, but civil liberties concerns over bias, oversight, and regulation are rising.
Meta is acquiring Singapore‑based AI agent startup Manus for over $2 billion to bolster its AI agent capabilities and integrate autonomous task‑performing technology into its platforms amid intensifying competition.