VoIP Extreme Free Call To PC To PC Via Router By SuDHiR
Jemima Torguson
Technology is not automatically accepted by users if it is difficult or confusing to interface with. DDS will discuss the concepts behind a superior user experience research and design process that enables increased adoption of new technologies like data analytics and AI.
This session will provide an overview of the data integration layer (DIL), including the purpose in providing ubiquitous access to DoD data and the necessary cultural, procedural and technical change that has to occur to achieve it.
The DoD Combat Support Agencies work across boundaries to support both the IC and Warfighters. This talk will provide a high-level overview of the current challenges Combat Support Agencies face to achieving decision advantage.
How do we get to a data-driven and AI-enabled Department? How do we ensure our warfighters are enabled to successfully deter in any domain, and if necessary, decisively win in any conflict? Discover how shared investments can unlock economies of scale as we drive towards a collaborative data empowerment for a more responsible, inclusive, and efficient AI landscape.
Mission Command applications are the class of software products that allow automation of Battle Management, Command and Control functions for Strategic and Operational commanders. This session will highlight the capability of the currently fielded applications and identify where gaps exist in the process.
As CDAO looks ahead, it discusses the strategy, opportunities and challenges for enabling the DoD to become more self-sufficient in solving their own data and analytic needs. Learn more about how CDAO intends to partner with customers to ensure they are connected to the full suite of CDAO service offerings that meet their needs.
Attracting and retaining skilled professionals proficient in AI has become a critical challenge for organizations worldwide. This panel will examine the strategies, insights, and innovative approaches drawn from industry experiences to effectively recruit and retain AI talent. We will explore case studies, best practices, and lessons learned from leading companies that have successfully navigated the complexities of talent acquisition and retention in the dynamic landscape of AI.
Discussion of the ethical challenges associated with developing and using LLMs in the national security context, including concerns related to bias, hallucination, and interpretability, and how those challenges differ from ethical challenges associated with traditional AI. The session will also touch on opportunities for industry and government to work together to advance the ethical use of LLMs.
This talk will highlight research challenges cutting across Army, Navy, and Air Force use cases for large pre-trained models. These include research on multimodal models, knowledge distillation for deployment at the edge, continual learning, model composition, advanced reasoning capability, among others. We will also argue for a multi-tiered DoD compute infrastructure for AI R&D.
CDAO will share their lessons learned from testing critical AI systems, such as the measurement of data drift, testing of AI integration with legacy systems, and continuous testing. They will also provide with their insights on AI test and evaluation moving forward, including avenues for industry and DoD collaboration.
As DoD systems become integrated with AI and autonomy capabilities, the question of novel attack surfaces and vulnerabilities arises. While adversarial AI has become a topic of great interest in recent years, much of the existing work within the field of adversarial AI has been done within academia and research.
This panel emphasizes the importance of Human Systems Integration (HSI) Test and Evaluation (T&E) throughout the lifecycle of an AI-enabled system, advocating for its implementation early, often, and always. Traditional HSI T&E data is usually captured through discrete experiments, an approach not well-suited for the automated, continuous testing required for AI/ML models. The panel will discuss (1) the challenges in instrumenting HSI-relevant data capture, (2) strategies and methodologies for integrating HSI into automated, real-time testing environments, and (3) innovative measures that utilize real-time user inputs such as search queries, tone of voice, response latency, and sentiment analysis.
We analyze the present and future impacts of Large Language Models (LLMs) on cybersecurity and cyber operations through both theory and real-world examples. This session will explore how large language models are creating new capabilities and risks for cyber threat actors and defenders alike.
DoD dev activities benefit from direct access to mission data during development efforts. Discussion Questions: What are models for acquiring mission data in a non-interference manor? What are the pros, cons, and best practices for supplementing or replacing mission data with synthetic data? What resources exist to host mission data in test or other dev environments?
A collaborative discusion around the challenges associated with Agile authorizations and AI. Outline the opportunities and exemplars. Lead to identification of interest in in participation of a CDAO/Industry AO round table being formed to continue the collaborations.
The United States works closely with allies and partners to apply existing international rules and norms and develop a common set of understandings among nations guiding the responsible use of AI in defense. This panel provides the opportunity to promote order and stability in the midst of global technological change. The United States has been a global leader in responsible military use of AI and autonomy, with the Department of Defense championing ethical principles and policies on AI and autonomy for over a decade. Among various national and international efforts, the United States, together with 46 nations, endorsed the Political Declaration on Responsible Military Use of Artificial Intelligence and Autonomy in November 2023, providing a normative framework addressing the use of these capabilities in the military domain. Given the significance of responsible AI in defense and the importance of addressing risks and concerns globally, the internationally focused session at the Symposium will be focused on these critical global efforts to adopt and implement responsible AI in defense. This panel will provide various country perspectives on the development, adoption, and implementation of principles and good practices on responsible AI, including multilateral efforts related to the Political Declaration on the Responsible Military Use of Artificial Intelligence and Autonomy.
This workshop is a series of industry and government presentations of selected LLM use cases. The creators will present their case and then discuss with the audience. The goal is to begin to develop acceptability criteria that will be presented on the last day of the symposium.
To maximize mission impact and ROI, the DoD needs to increase the access and exchange of data across the Components, theaters, and domains. Discussion Questions: How can DoD SMEs and industry identify the right specifications to standardize the creation and deployment of increased API endpoint services across the diversity of DoD data types and mission domains? What are best practices in creating flexible, resilient, future proofing API services? What is the right documentation and reference resources to support API services?
DoD needs to take a flexible approach to digital personnel, both uniformed and civilian, to take full advantage of the available digital talent pool. This conversation will explore the problem and solution spaces of building teams out of different military and civilian components.
What does it mean to have data-driven management, and how does that look in the DoD? In this session, our Business Analytics Directorate will show the capabilities of Pulse and how it has changed DoD management.
The DoD data science community collectively and at the Component-level are lacking the appropriate access to sandbox environments and compute resources. Discussion Questions: What is an average or reasonable compute resource allocation for data science use and is there a stratification across experience levels or mission applications? What are the right specifications for data science sandbox environments and where do they exist in the DoD today? Are data science environments best allocated from onprem distributive computing clusters or cloud-based resources?
This session provides an overview of Other Transaction Authority (OTA) mechanism available to engage small businesses in the procurement of innovative capabilities. The Tradewindai.com portal will be discussed and demonstrated to support the discussion.
LLMs are an imperfect tool that has incredible potential but also self-evident shortcomings. Leaders, Industry and scientists must weigh the benefits and risks when deciding how to use this technology. This session is a kickoff to the conference-long LLM track of the symposium and highly recommended for all LLM workshop participants.
Time is of the essence to deliver AI-enabled and data-driven decision-making capabilities to the warfighter. In order to win, the Department must adopt design thinking and agile software development methodologies. CDAO is paving the way. Hear three TED talks where AW has successfully partnered with the Services to produce use case and user-driven prototypes that inform and shape what CDAO needs to deliver as an enabler of the Department.
Emerging technologies rely on a workforce that understands the importance of not only improving technology but also protecting it through actions such as reporting suspicious behaviors and activities. The presentation describes individual behaviors and work culture characteristics indicative of risk and threat from trusted insiders and unpacks reporting versus non-reporting behaviors. The presentation will also address cultural change strategies to build and maintain secure cyber landscapes through deliberate and ongoing engagement with people in ways that support the protection of our rapidly emerging technology spaces.
7fc3f7cf58