Welcome to the AI Training Series for Government Employees
Enhance your skills and knowledge of the rapidly evolving world of artificial intelligence (AI).
Our comprehensive training series, which meets the training requirements of the Artificial Intelligence Executive Order, is designed to inform and educate government employees at all levels, offering specialized tracks to meet the diverse needs of the government workforce. The TTS Centers of Excellence AI Community of Practice (AI CoP) conducted this AI Training in partnership with Stanford HAI, GWU Law School, Princeton CITP, Wilson Center, GSA OGP, and OMB.
Access the 2024 Training Series
The 2024 AI Training Series held live sessions in September and October 2024. The recorded sessions have been transformed into e-learning modules and are now available for government employees via USA Learning.
Technical Track
In partnership with Stanford Human-Centered AI, this track breaks down complex AI concepts into plain language, covering human-centered AI development, privacy and security concerns, and risk mitigation techniques.
| Navigating the AI Landscape | This course provides a comprehensive overview of AI, including the definition, theories of AI and machine learning, neural networks, narrow vs. general AI, gradient descent, use cases, and more. |
| Privacy & Security | This course covers how different social values around privacy, data ownership, and data creation will impact what AI technologies are possible today and what the future paths of innovation in AI will look like. |
| AI Safety & Robustness | The course looks at the considerations that AI developers must evaluate when designing AI systems for safety such as how to address biased inputs, navigate constantly evolving conditions, and address explainability issues. The course will look at how we can navigate all these risks and design the right parameters for safety. |
| Generative AI Fairness | This course provides an in-depth understanding of how biases embedded in data can lead generative AI models to make certain predictions that are systematically different across groups and how to assess algorithmic fairness for human-facing applications of generative AI. |
| HELM and Benchmarking Foundation Models | This course covers the importance of benchmarking AI models as a way to understand the capabilities of systems and promote transparency amongst model developers. The course will also discuss Stanford’s Holistic Evaluation of Language Models (HELM) benchmarking approach which serves as a model to evaluate language models. |
| Building and Training Foundation Models | This course provides more depth into what goes into building a foundation model. It discusses the challenges associated with training and the various stages of model development. |
| Training Cost-Effective Large Language Models | The cost of training large language models (LLM) is currently quite high with top of the line models costing billions of dollars to train. However, are costs expected to always remain that high? This course looks into the resources required to train LLMs and what could be done to make training them more cost effective. |
| Multimodal Foundation Models | Multimodal foundation models are capable of processing and generating both visual and textual information. This course will discuss how the development of these models could further expand generative AI capabilities and their potential downstream uses. |
Acquisition Track
In partnership with George Washington University Law School, these sessions cover the fundamentals of AI procurement to understanding risk management and ethics in AI acquisition.
| Buying AI: Government Contracts 101 | A foundational overview of basic federal procurement policies and requirements as they relate to AI, so that attendees have a better understanding of the goals and constraints of U.S. federal acquisition. |
| How Does AI Benefit the Federal Government? | Customer needs and satisfaction are a foundational underpinning of the U.S. procurement system. This session will discuss the ways in which AI may benefit “the business” of the U.S. federal government. |
| Risk Management & Ethics | An overview of the ethical considerations and risk mitigation measures essential to responsible AI acquisition. |
| Developing a Long-Term AI Acquisition Strategy | This session will focus on special considerations for a long-term AI acquisition strategy, such as ensuring explainability, how AI could harm the federal government and U.S. citizens, as well as statutory and policy compliance, including meeting U.S. government technical best practices. |
| National Security AI Acquisitions | AI acquisitions to advance national security agency missions and programs or those that enhance tech stacks require special considerations of existing and new requirements. This session will survey the landscape and outline key considerations to make sound decisions. |
| Data Privacy Considerations | An overview of general and AI procurement-specific data privacy considerations that are crucial to an effective and compliant procurement strategy, including mitigating the risks of harms associated with AI. |
| Compliance with AI-Related Regulations | The acquisition of AI is governed by a myriad of regulatory requirements. This session will help attendees identify and understand the key regulations that govern the purchase of this technology. |
Leadership and Policy Track
In partnership with the Center for Information Technology Policy (CITP) at Princeton University, these sessions explore AI policy development, ethical leadership, and strategic planning, ensuring leaders are well-prepared to handle the societal impacts of AI technologies.
| AI Foundations for Decision Makers | This session provides a comprehensive introduction to the science behind artificial intelligence, focusing on how AI systems work and their key technological features. It also equips decision-makers with the skills to identify misleading AI claims and distinguish them from genuine technological advances. |
| AI Strategies and Insights | This session looks at how the government can leverage AI for enhanced decision-making and operational efficiency. The session also introduces frameworks for the ethical adoption and implementation of AI tools in public sector initiatives. |
| Risk & Mitigation | This course examines the potential risks AI poses, such as bias and privacy concerns. The session offers practical strategies for mitigating these risks to ensure responsible and secure AI implementation. |
| AI Auditing | This session guides government leaders on how to audit AI systems to ensure they adhere to performance, safety, and ethical standards. The session emphasizes the critical role of rigorous assessments in maintaining accountability and trust in AI technologies within the public sector. |
| Future Trends in AI | This course examines emerging trends in artificial intelligence and provides insights on how to anticipate and prepare for future developments in the field. |
| AI & Security | This course explores the vulnerabilities of AI systems to potential attacks and methods for protecting against these threats. The session also highlights efforts to develop and identify AI that is reliable, safe, and trustworthy. |
Access the 2023 Trainings Series
The recordings of the 2023 and 2024 sessions are available to government employees on the AI CoP USDA Connect page.