The mandate of the National Security Commission on Artificial Intelligence (NSCAI) is to make recommendations to the President and Congress to “advance the development of artificial intelligence, machine learning, and associated technologies to comprehensively address the national security and defense needs of the United States.”

This Final Report presents the NSCAI’s strategy for winning the artificial intelligence era. The 16 chapters in the Main Report provide topline conclusions and recommendations. The accompanying Blueprints for Action outline more detailed steps that the U.S. Government should take to implement the recommendations.

Document Archive

Final Report of the National Security Commission on Artificial Intelligence, July 2021 [756 Pages, 14MB]

Loading...

Taking too long?

Reload document

| Open in new tab

Download [14.01 MB]

The post Final Report of the National Security Commission on Artificial Intelligence, March 2021 first appeared on The Black Vault.

Artificial intelligence and machine learning (AI/ML) is a next-generation computer capability that holds the potential for changing everything from how people live and work to how wars are fought and won. The Combat Capabilities Development Command Chemical Biological Center (DEVCOM CBC) is keeping pace with this computing revolution through its Grand Challenge Program, beginning with three pilot projects and a workshop to recruit more.

Video Archive

The post U.S. Army Lab Explores AI/ML Potential in Development of Chemical Biological Defense Solutions first appeared on The Black Vault.

Signaling a major leap forward for national defense in the digital age, the Air Force flew with artificial intelligence as a working aircrew member onboard a military aircraft for the first time Dec. 15.

The AI algorithm, known as ARTUµ, flew with the pilot, U.S. Air Force Maj. “Vudu”, on a U-2 Dragon Lady assigned to the 9th Reconnaissance Wing at Beale Air Force Base. Air Combat Command’s U-2 Federal Laboratory researchers developed ARTUµ and trained it to execute specific in-flight tasks that otherwise would be done by the pilot.

The test flight was the result of years of concerted effort within the Air Force to apply cutting-edge technology to military operations as it competes with other world powers in the digital age.

“ARTUµ’s groundbreaking flight culminates our three-year journey to becoming a digital force,” said Dr. William Roper, assistant secretary of the Air Force for acquisition, technology and logistics. “Putting AI safely in command of a U.S. military system for the first time ushers in a new age of human-machine teaming and algorithmic competition. Failing to realize AI’s full potential will mean ceding decision advantage to our adversaries.”

During this flight, ARTUµ was responsible for sensor employment and tactical navigation, while the pilot flew the aircraft and coordinated with the AI on sensor operation. Together, they flew a reconnaissance mission during a simulated missile strike. ARTUµ’s primary responsibility was finding enemy launchers while the pilot was on the lookout for threatening aircraft, both sharing the U-2’s radar.

The flight was part of a precisely constructed scenario which pitted the AI against another dynamic computer algorithm in order to prove the new technology.

“We know that in order to fight and win in a future conflict with a peer adversary, we must have a decisive digital advantage,” said Air Force Chief of Staff Gen. Charles Q. Brown, Jr. “AI will play a critical role in achieving that edge, so I’m incredibly proud of what the team accomplished. We must accelerate change and that only happens when our Airmen push the limits of what we thought was possible.”

After takeoff, the sensor control was positively handed-off to ARTUµ who then manipulated the sensor, based on insight previously learned from over a half-million computer simulated training iterations. The pilot and AI successfully teamed to share the sensor and achieve the mission objectives.

The U-2 Federal Laboratory designed this AI technology to be easily transferable to other systems and plan to further refine the technology. Today’s flight provided invaluable data for not only the team to learn from, but also ARTUµ.

“Blending expertise of a pilot with capabilities of machine learning, this historic flight directly answers the National Defense Strategy’s call to invest in autonomous systems,” said Secretary of the Air Force Barbara Barrett. “Innovations in artificial intelligence will transform both the air and space domains.”

The U-2 Federal Laboratory is a 15 U.S.C. compliant organization established to bring together a “confluence of warfighter, developer, and acquirer” vertically-integrated under the same operational roof. The lab has developed and been approved by the National Institute of Standards and Technology to establish the 20th Laboratory Accreditation Program in the federal government. It promotes “edge development” – a concept to develop new software integration on operational systems in a bounded, safe environment.

The historic flight with AI comes just two months after the U-2 Federal Laboratory team updated inflight software for the first time during a U-2 training mission. The team leveraged the open-source container-orchestration software Kubernetes, another military first.

The post AI Copilot: Air Force Achieves First Military Flight with Artificial Intelligence first appeared on The Black Vault.

The following documents were obtained under the Freedom of Information Act (FOIA) from the Central Intelligence Agency (CIA). They were requested in November of 2016, and finally released in April of 2018.

They profile the research done by the USSR on Artificial Intelligence in the mid 1960s.

Although this is considered one “document” by the CIA (as indicated by the MORI ID number on the top left of the pages), it does comprise of multiple records regarding USSR Artificial Intelligence research.

Document Archive

 CIA Scientific Intelligence Reports on Artificial Intelligence Research in the USSR, September 1964 [68 Pages, 10.3MB]

Loading...

Taking too long?

Reload document

| Open in new tab

Download [9.74 MB]

The post CIA Scientific Intelligence Reports on Artificial Intelligence Research in the USSR, September 1964 first appeared on The Black Vault.

The following documents were received that pertain to research into the development and use of Autonomous Robots.

Autonomous Robots

In chronological order from newest to oldest

Cooperative Autonomous Robots for Reconnaissance, June 3, 2009 [9 Pages, 25  KB] – Collaborating mobile robots equipped with WiFi transceivers are configured as a mobile ad-hoc network. Algorithms are developed to take advantage of the distributed processing capability inherent to multi-agent systems. The focus of this study was to determine the optimal amount of communication which allows the robots to share a sufficiently detailed global map, while keeping their processing time and energy usage to a minimum. A hardware testbed is described, which will be used to examine these trade-offs in an indoor laboratory-scale test area.

Lethality and Autonomous Robots: An Ethical Stance, 2007 [4 Pages, 196 KB] -This paper addresses a difficult issue confronting the designers of intelligent robotic systems: their potential use of lethality in warfare. To fully understand the consequences of the deployment of autonomous machines capable of taking human life under military doctrine and tactics, a systematic ethical evaluation needs to be conducted to guide users (e.g., warfighters), system designers, policy makers, and commanders regarding the intended future use of this technology. This study needs to be conducted prior to the deployment of these systems, not as an afterthought. Toward that end, a 3-year research effort on this topic is being conducted at the Georgia Institute of Technology for the Army Research Office, of which the authors are currently in the first year. Two topics are being investigated: What is acceptable?, and What can be done? A survey is being conducted on the use of lethality by autonomous systems. The survey investigates the points of view of various demographic groups on this issue, including the public, robotics researchers, policy makers, and military personnel. The authors also are designing a computational implementation of an ethical code within an existing autonomous robotic system (i.e., an artificial conscience) that will be able to govern an autonomous system’s behavior in a manner consistent with the rules and laws of war. This paper describes the survey’s design and administration. The independent variables used for the survey are as follows: (1) community type; (2) level of authority; (3) demographic variables, such as age, gender, level of education, etc.; and (4) the extent of participants’ knowledge of robots and their capabilities. In addition to finding out what the terms of acceptance are for using lethal robots in warfare, the authors would like to see if, and how, the level of acceptance varies among the different community types, according to certain demographics factors, and for the three levels of autonomy.

If Our Robots Are So Smart, Why Aren’t We All Rich? (The Challenges of Integrating Autonomous Robots) November 8, 1999 [43 Pages, 997 KB] – Presentation of military research efforts into the development and applications of robotic systems.

A Technique for Coordinating Autonomous Robots, April 12, 1986 [7 Pages, 910 KB] – This paper describes a technique for coordinating the subsystems of autonomous robots which takes advantage of a distributed blackboard mechanism and a high degree of functional distribution between subsystems to minimize communications and simplify the interfaces. Distributed blackboard memory contains a world model which represents knowledge about itself and its surroundings as collections of objects important to the task and the relations between them. Objects or instances are represented as lists of object-attribute- value-accuracy-confidence-timestamp tuples which are organized into a class tree with inheritance properties and active functions.

Current Technical Research Issues of Autonomous Robots Employed in Combat, September 1984 [6 Pages, 200 KB] – The recent upsurge in interest in autonomous robots for combat applications has focused considerable attention on several of the obvious technical issues (e.g. target recognition, autonomous navigation, route planning) . However, several technical issues exist which remain unapproached and, in some cases, even unacknowledged by the robotics community. This paper explores three such issues: (1) robot fault tolerance, (2) robot security and (3) multi-robot coordination. These issues are discussed in terms of the technology limitations and the research issues associated with those limitations. A common message which occurs several times during this discussion denotes the importance in modular implementation and well defined interfaces between subsystems in the development of autonomous combat robots.

Autonomous Satellites

Autonomous Mission Management for Satellite Systems, August 15, 2015 [67 Pages, 7.6MB] – It’s almost humorous they would send this document to me, as a response to a FOIA request. You will note, it is nearly entirely redacted, with the exception of the footer, and the distribution list.

Abstract: Satellite intelligence information is being used increasingly for real-time operations. This requires satellites that can be quickly tasked for new objectives and that can respond to opportunistic situations and external threats. Unfortunately, today’s satellites rely on ground operators to control them. This means there is a large delay between when a new task is requested, or a new threat or target of opportunity is identified, and the response by the satellite. TRACLabs proposes the The HAMMER system, which is designed to operate on-board a satellite, controlling the satellite”s goal-oriented activities and responding to threats even when it is not in communication with the ground or when time constraints require immediate response to threats and faults. The HAMMER system attempts to meet mission objectives even in the face of threats and faults. HAMMER prioritizes multiple, competing user goals and requests and determines a satisfactory ordering of satellite tasks to conserve resources and maximize capability. HAMMER autonomously executes the tasks by issuing commands to the host satellite system. BENEFIT: We believe that HAMMER provides the following benefits to the Air Force: 1) Rapid and continuous response: does not require ground to be in the loop; 2) Self-defense responses to threats; 3) Immediate mission replanning due to disturbances or opportunities; 4) Guaranteed safety (from modeled threats, given defensive response capabilities); 5) High-level tasking of spacecraft: planning and execution framework fills in fine-grain command details; and 6) Increased spacecraft productivity: more observations, more tasks accomplished. This work has commercial applications in enabling the automation of vehicles such as satellites, unmanned vehicles (e.g., unmanned air vehicles (UAVs), unmanned ground vehicles (UGVs), unmanned surface vehicles (USVs), etc.), and manned vehicles (e.g., aircraft, spacecraft, ships, etc.). This work also has commercial applications in the oil and gas industry and in the industrial robotics industry.

Autonomous Mission Management for Satellite Systems, January 2002 [67 Pages, 3.9MB] – This report summarizes mobile foundations Phase I SBIR project entitled An Automated Tool to Enable the Distributed Operations of Air Force Satellites. The overall goal of the project was to proved the feasibility of enhancing US Air Force space operations through the use of advanced automation to provide distributed situational awareness. Such an approach will help the Air Force meet the vision of Next Generation Space Operations laid out in the Air Force Space Command (AFSPC) Strategic Master Plan. This report documents the human effectiveness and systems analyses mobile foundations used as a basis for its proof-of-concept prototype. The report also describes the software prototype (called FASAT, Fast Access Situational Awareness) that mobile foundations developed and demonstrated to prove the feasibility of its approach to developing a next-generations distributed operations system.

The post Autonomous Technology / Robots first appeared on The Black Vault.

Artificial intelligence (AI) is the intelligence exhibited by machines or software. It is also the name of the academic field of study which studies how to create computers and computer software that are capable of intelligent behavior. Major AI researchers and textbooks define this field as “the study and design of intelligent agents”, in which an intelligent agent is a system that perceives its environment and takes actions that maximize its chances of success. John McCarthy, who coined the term in 1955, defines it as “the science and engineering of making intelligent machines”.

The following are documents relating to artificial intelligence. Since the evolution of technology moves very quickly, I have started the archive to appear chronologically (then alphabetically) to differentiate the difference is scientific knowledge and technological capabilities over the decades.

Declassified Documents

 Some Problems and Proposals for Knowledge Representation, 06 August 1987 [44 Pages, 5.5MB] – Knowledge representation is widely regarded as a central problem in artificial intelligence. However, there appears to be no convergence of opinion as to the form a knowledge representation system should take, the principles it should embody, or even what its goal should be. While progress in the past decade has led to a number of interesting theories and useful programming formalisms, this research has also raised doubts about the adequacy of the foundations of many of these ideas. This paper presents some observations about the knowledge representation schemes now in common use. Some of these observations are critiques of these schemes, or extensions of critiques made by others. To remedy some of these problems, a new theory of knowledge representation is proposed. The theory attempts to encompass representational ideas that have emerged from different schools of thought, in particular from work in semantic networks, frames, frame semantics, and Conceptual Dependency.

The post Artificial Intelligence first appeared on The Black Vault.