The Chinese central government has led PPP development of its “Sharp Eyes” surveillance system.3 By intertwining digital commerce with public safety, China has created an unprecedented surveillance apparatus with near limitless opportunities for machine learning and analytics to process, categorize, and contextualize information for human operators. This surveillance system model now exported around the world as “Safe City” solutions challenges the Western notion of privacy and human rights when employed against targeted population groups like the Muslim Uighurs in western China.
In light of this range of applications, the definition of “situational awareness” remains somewhat ambiguous and nonpractical. For the purpose of this paper, SA refers back to the foundational definition espoused by Dr. Endsley and refers to the real‐time presentation of pertinent information to a human operator to inform subsequent action. The geographic domain is relevant only in so much as its relevance to the human operator in question. Similarly, historic information and trends are relevant only in as much as they apply to the real‐time context of the operator. Multiple operators may be involved in a single event, and the SA platform must consider the perspective and context of each in order to achieve its intended purpose.
2.3 Putting Situational Awareness in Context: First Responders
Although much literature has been written on SA concepts in the aerospace, military, and maritime domains, the proliferation of Internet of Things (IoT) devices and advancements in machine vision and AI have enabled the democratization of SA capabilities. Under the banner of “smart cities,” municipalities have begun implementing static surveillance capabilities and outfitting first responders with mobile and body‐worn devices that act as both a sensor and a means of improving SA. By some estimates, the market for surveillance equipment will reach $77B by 2023.4 This explosion in sensors has led to increased public safety expectations, as well as greater scrutiny over the actions taken by first responders.
To meet these expectations, law enforcement agencies in particular employ a variety of surveillance tools to achieve awareness of events occurring in the geographic domain under their authority. These tools include closed‐circuit television (CCTV) cameras; license plate readers; and chemical, biological, radiation, nuclear, and explosive (CBRNE) sensors. Historically compartmented information warehouses containing criminal histories, emergency calls, use of force logs, and similar are increasingly being fused and made available for real‐time search. Moreover, noncriminal information ranging from social media and other open‐source datasets to credit histories and other quasi‐public records are increasingly accessible to provide context to an event. The use of such noncriminal records to assist law enforcement is often vigorously contested and will be addressed later in this chapter, but regardless of a particular agency's implementation, today's challenge remains a big data problem. In other words, identifying the particular set of information relevant to an event is paramount; with few exceptions, the requisite data points to improve an officer's SA are available.
Complicating the analysis and dissemination of pertinent information to SA are the layers of information security policies applied to the first responder community. For example, law enforcement agencies in the United States must adhere to the Criminal Justice Information Standards established by the Federal Bureau of Investigation.5 These standards mandate, among other requirements, that anyone accessing law enforcement data first authenticate themselves as a qualified operator and further establish a need to know the information requested. These regulations are often further restricted by agency‐specific policies, such as preventing the disclosure of information pertaining to active cases to anyone not specifically associated with the case in question. Such policies and regulations were generally enacted and expanded in the wake of inadvertent (or deliberate) misuse of information over many decades. Few contemplated the ramifications on nonhuman actors, such as the potential of AI, and fewer still considered how persistent access to such information may contribute to real‐time SA platforms charged with improving the safety and effectiveness of modern‐day first responders.
It is in this context that the demands on first responders to employ SA platforms for decision support are being placed. With this comes a myriad of HSI concerns, ranging from the physical real estate available to first responders to interact with SA platforms to the means by which this complex set of information can be presented. Underpinning all considerations is the paramount importance of officer safety and the need to understand the operator's context in order to establish information relevance and right to know.
2.4 Deep Dive on Human Interface Considerations
With the advent of IoT sensors and significant increases in capabilities for both connectivity and storage, big data has become the prime dependency for many new technologies and solutions, especially SA. In public safety, and more particularly with first responders, the sheer breadth of information available is overwhelming. Designing human system interfaces that can retrieve, parse, and organize relevant data based on real‐time activities and events, as well as present it in a meaningful, concise, and unintrusive (yet attentive) way, is a defining challenge.
At its core, public safety focused SA is predicated on alerting to noteworthy events in real time while increasing the knowledge and expanding the experience of responding personnel by drawing upon all pertinent historical, concurrent, and predictive information available to the agency. With a primary focus on officer safety, users of this system only have a few minutes upon being notified of the event to ingest the relevant data, make a determination on tactics, and adjust their response accordingly. This is all while they are also driving, communicating with dispatch, and coordinating with colleagues and supervisors. As such, the intelligence generated and presented must offer substantive benefits as rapidly and concisely as possible. The immediate goal of all first responders is to protect life, and much of the data available to police departments can support key areas such as subject identification, threat assessment, and response tactics, all of which greatly enhance SA and help to keep everyone safe.
Machine‐assisted data retrieval, organization, and presentation not only improve the safety of all those involved, but it supports officer decision making by informing them of supplementary details and historical activities and actions. These characteristics are unique to every call for service, and a better understanding of them within the context of the current interaction is invaluable. However, the same mechanisms that collate the appropriate information must also exclude the rest. Considering the highly mobile nature of first responders and the inherent limitations of portable hardware in a public safety setting, it is not practical to expose all associated data, even if it could potentially be relevant in some ancillary contexts. Conversely, ignoring that information has its own tangible detriments, most notably, indicating an incorrect narrative to responding personnel that causes them to make poor judgments that have lasting impacts.
Computers have a unique ability to project truth, regardless of the quality and completeness of the underlying data. This “machine heuristic”6 easily combines with algorithmic bias, which can corrupt the decision‐making process for first responders with little apparency. As an example, domestic violence incidents are some of the most volatile and dangerous in policing and one in which historical context can greatly sharpen the officer's picture of the situation. Here, the prioritization of arrest history of the involved individuals would seem prudent, since it can assist with the identification of the primary aggressor. However, if that lessens the visibility, or excludes completely, non‐arrest situations where the incident was resolved without enforcement action, it can skew