Center for Strategic Assessment and forecasts

Autonomous non-profit organization

Home / Defence and security / New in the military / Articles
The Pentagon replace commanders by thinking machines
Material posted: Publication date: 22-09-2019
The situation in the world, is very similar to the one in the famous film "Seventeen moments of spring" was described by the Wehrmacht General, who was riding in the same train with Stirlitz in the period after the Second world war. In his evaluations, was the role of politicians and diplomats in the early wars, American views on their conduct, the responsibility of aggressors for crimes against humanity, and much more, which greatly resonates with today.

Despite the progress made by humanity, everything remains virtually unchanged, and the best example is the overseas establishment that claims to hegemony and the Prosecutor of the planet and making this all possible and even beyond the mind. The United States is very actively arm themselves and create new systems and means of armed struggle. In addition, for the military Department of America, the development and implementation of hardware and software of artificial intelligence (AI) in all areas of military construction, especially in exploration, is one of the priorities.

September 5 speaking at the conference on intelligence and national security, Deputy Minister of defense for intelligence Kari bingen noted the success of the Pentagon in the development and implementation of systems and tools of AI, created during implementation of the project "Maven" (Maven Project) in the activities of the intelligence units. According to her, the innovation will help to change many processes in exploration for the better.

"We need changes to work processes, efficiency and transformation of the professional culture as a result of the introduction of artificial intelligence into the activities of specialists. It's actually what we expect in the next year", – said the Deputy Minister, in a speech at the conference on intelligence and national security, held in Washington on 4-5 September.


In early February of this year, the military establishment of America has published "a summary of the strategy of the Ministry of defence on artificial intelligence for 2018".

According to experts, the new doctrine, among other things, aimed at countering Russia and China in this field. In the Pentagon assert that investments of other countries in artificial intelligence for military purposes could undermine the technological and operational advantages of the United States, as well as to destabilize the free and open international order.

This document lays the Foundation structure interaction MO in the process of military technology creation of AI as from the point of view of the final application, and from the point of view of organizational issues when introducing them. According to experts of the Pentagon, the technology of AI could radically change the order of preparation and decision-making in various fields of military activities. The authors of the Strategy AI (FIC) believe that MO is obliged to join in the arms race and to determine the scope of the use of new technologies, as they have already started to implement in Russia and China, including in areas that are critical in terms of ethics.

The framers of the SSI noted that the introduction of AI is an iterative process of development of new technologies that goes beyond the specific ideas presented in the document. However, it is formulated a few very specific suggestions for technology use AI in MO USA. The main and closest in time of realization of the authors of the Strategy consider so-called predictive maintenance and logistics (predictive maintenance and supply), the essence of which is that with the constant monitoring of arms and military equipment (AME) can reveal signs of aging before they actually become unfit for use, and more efficient use of repair resources. This should reduce operating costs and improve the ratio of combat-ready forces to their total number, the so-called ratio of the operational voltage.

The document mentions the use of AI to enhance shared situational awareness and recognition of emerging and non-obvious for the user of the dangers in all areas of their activities, starting in cyberspace and ending with the classic battlefield. These dangers can occur because of potential malfunctions and defects, and because of the actions of the enemy. Raising awareness should, as planned by the Pentagon to improve the quality of situation assessment and improve the accuracy of military action. The Pentagon also hopes to predict unsafe situations to improve decisions and enhance the safety of both military and protected objects, for example, the critical infrastructure of the United States.

Anticipates the paper and what the authors call decision support. Translated into comprehensible categories, this means the calculation of plans of action in a given situation. In the case of equipment maintenance is to find the optimal strategies for the maintenance and or evaluation of actions, and make recommendations to the operator or full automation control repair units.

Regarding automation, as such, the authors write very cautious and limited only to the most General formulations. Technologies associated with AI, need to automate "routine, manual and unskilled labor" to improve productivity, to reduce the likelihood of error determined by the human factor, and to allow operators to concentrate on higher-level tasks. All together it should turn the Pentagon into a major consumer of funds AI, and to improve the efficiency of internal processes in the Ministry.

For the aforementioned transformation of the defense Ministry proposes to create an additional structure, the joint center for artificial intelligence (Artificial Intelligence Joint Center III), whose task will be the integration of AI technologies into the daily activities of the military. The intent of the Pentagon, III needs to perform several functions.

Organizationally, the center should coordinate the implementation of appropriate technologies in all spheres of military activities. This means that it will act as a bridge between the end users of AI systems and create their researchers, allowing you to immediately implement new developments in military context. The second goal of creating a unified centre is to remove administrative and procedural obstacles for the use of the same practices in different types of troops.

Create III is a logical step in the way of concentration of competent personnel in the AI within the single structure of the Pentagon, which, on the one hand, will coordinate the implementation of AI in the armed forces, and on the other be responsible for communication in this area with allies and scientific-technical centers of fundamental and applied research such as the Office of advanced studies of MO (DARPA) or University laboratories.

It is important to note that III conceived primarily as an Executive structure, the primary objectives of which are the development, evaluation and, if successful, tests on the battlefield – the wide-scale adoption of prototypes produced systems and tools AI. While communication skills are considered essential to this task. In this framework it is the responsibility III will include the standardization of development systems and tools of AI, including their development and application.

For a quick set of necessary administrative skills in the field of creation and application of AI, the Pentagon intends to rely on cooperation with industrial partners, which should ensure the reduction of time of tender procedures. It is also planned to involve the academic community and private firms to solve the serious problems not related to military specifics. Thus experts suggest MO not only solve the challenges they face, but also to ensure the engagement of qualified professionals.

In the document in a separate paragraph talks about the interaction with the scientific community by funding research, both through traditional channels such as DARPA, and through new funds, created to Finance research in the field of AI. Here the Pentagon is not only interested in the new procedural possibilities, but the answers to the question how to create a safe for the military use of AI that can be explained by the specialist. In the same context, the defense establishment is committed to the ethical principles of using AI and to fight for their international adoption.

It is significant that in the motivational part of the drafters of THESE underscore the need to include in the competition for technology AI that likely opponents are already developing and introducing appropriate technologies with far-reaching geopolitical goals, conflicting values of the US and its allies. Therefore, the document repeatedly referred to the ethics and moral values.

Apparently, the Pentagon formed the opinion that the introduction of technology AI has neither the necessary financial resources or organizational structures. Therefore, the strategy envisages, in addition to actually implementing the technology, additional active funding, involvement of personnel from civil and scientific sectors, as well as the creation of a specialized internal departments in AI, which must manage the introduction of technology and partly to transform not only the implementation process, but also the Pentagon, increasing its effectiveness not only through technical innovation but also through the application of modern management practices.


As reported by some media, the American military authorities to increase spending on secret projects, the purpose of which is the use of software and hardware, AI is not only to anticipate a nuclear missile strike and identify the location of mobile launchers, but also for control of nuclear forces and use their weapons. Information about these developments has always been extremely scarce and was of a very fragmentary and rather unreliable character. However, in the middle of last year Reuter Agency, citing informed sources said that such operations are conducted in MO USA.

According to us officials, whose names are not disclosed in the United States implemented several programs classified as "secret", in which it is supposed to create AI systems for enhanced protection of the U.S. from potential nuclear missile attack.

If scientific studies are successful, the developed computer system will be able to think independently, quickly and accurately people to gather a lot of data, including satellite imagery, and finding the signs of an imminent missile launch.

In advance of receiving such information, the U.S. government will be able to take diplomatic steps to resolve the situation, and in the case of the inevitability of the attack, the military will have more time to try to destroy missiles prior to launch or catch.

"We must do everything in our power to locate the missile before it launches and try to obstruct her start with the land," – said one of interlocutors of the Agency.

Several US officials and budget documents, reporters learned that the administration of President Donald trump has proposed a more than threefold increase in funding for such developments and to allocate 83 million dollars. only one of the programs on missile defense with the help of artificial intelligence. The increase shows the growing importance for U.S. research in the field of creating AI systems for missile defense in conditions of growing military power of Russia and the alleged continued nuclear threats from North Korea.

"Using artificial intelligence and machine learning to find a needle in a haystack," said former Deputy Secretary of defense Bob Wark (Bob Work), tightly worked on the topic of military robots and technology AI. Concrete projects he did not mention.

One of the sources told Reuters that the emphasis of the programs at the AI of the Pentagon is to protect from Pyongyang. In Washington's growing anxiety about the ongoing North Korea's development of mobile missile that can hide in tunnels or forests or caves.

Although the military AI projects are kept secret, the Pentagon made no secret of his interest in artificial intelligence technologies. In particular, the broad media coverage received "Maven Project" (Maven Project), involving the use of AI to analyze images from the UAV and the recognition of people in images. Involved in the project Google in June 2018, was forced to announce the termination of cooperation with the U.S. military because of dissatisfaction of their employees, who considered this work a violation of human rights.

Details about the Pentagon on the creation of AI systems to detect potential missile threats and mobile launchers are extremely scarce, but Reuters has learned that the first prototypes are already being tested by the military.

According to the Agency, the draft, which is in the territory of Washington, involved both military experts and private researchers. Scientists are adapting for our purposes the technological advances developed by commercial companies, financed by venture capital Fund In-Q-Tel, associated with the CIA.

But the same Wark recently warned the Pentagon that the widespread use of artificial intelligence in the military sphere, including in the analysis the need of the use of nuclear weapons, could provoke a nuclear war. In conversation with a representative of the online magazine Breaking Defense he said that the use of artificial intelligence systems in the nuclear field could lead to rash decisions about the nuclear strike.

According to Wark, such systems can incorrectly assess risk factors and when certain parameters and indicators to launch Intercontinental ballistic missiles that could lead to a "disastrous scenario." "Imagine that the artificial intelligence system in the command center of Chinese troops, analyzing operational data, gives the signal that the US is preparing to attack China and recommends that a pre-emptive strike" – quoted by Wark edition.

Former Deputy head of the Pentagon as an example, has also led the Russian system "Perimeter", which the West calls the "Dead hand". According to him, this system can take seismic activity for the signals from nuclear explosions, to send the request to the appropriate headquarters and in the absence of a response to give a command to launch ballistic missiles.

Wark does not exclude benefit from the introduction of high technology in the military sphere, but calls to restrict the use of artificial intelligence in the control and management of nuclear weapons. But the U.S. military does not heed any warnings and continues to embed technology in all spheres of military activities and replace her.

In conclusion, I want to remember the dialogue shtirlitsa with the Wehrmacht General who, describing the Americans and their attitude to the war, said: "These idiots ruin their own equipment, they think they can win a war by bombing alone. They will increase their technical prowess and drown in it. It will destroy them like rust. They think they can do anything".

Vladimir Ivanov


Tags: assessment , USA

RELATED MATERIALS: Defence and security