Facial-recognition algorithms from Los Angeles startup TrueFace are good enough that the US Air Force uses them to speed security checks at base entrances. But CEO Shaun Moore says he's facing a new question: How good is TrueFace's technology when people are wearing face masks? "It's something we don't know yet because it's not been deployed in that environment," Moore says. His engineers are testing their technology on masked faces and are hurriedly gathering images of masked faces to tune their machine-learning algorithms for pandemic times. Facial recognition has become more widespread and accurate in recent years, as an artificial intelligence technology called deep learning made computers much better at interpreting images.

The cybersecurity skills shortage continues to plague organizations across regions, markets and sectors, and the government sector is no exception. According to (ISC)2, there are only enough cybersecurity pros to fill about 60% of the jobs that are currently open -- which means the workforce will need to grow by roughly 145% to just meet the current global demand. The Government Accountability Office states that the federal government needs a qualified, well-trained cybersecurity workforce to protect vital IT systems, and one senior cybersecurity official at the Department of Homeland Security has described the talent gap as a national security issue. The scarcity of such workers is one reason why securing federal systems is on GAO's High Risk list. Given this situation, chief information security officers who are looking for ways to make their existing resources more effective can make great use of automation and artificial intelligence to supplement and enhance their workforce.

We are seeing AI technologies increasingly deployed across many parts of society. Around the globe, governments are rushing to mobilize vast amounts of capital to invest into AI innovation. The COVID-19 pandemic prompts us to rethink what is considered high- or low-skill work. Whose skills, whose labor and whose hours, exactly, are of value to society? What and who do we value and deem essential, and how do we compensate these workers (e.g., care work or teaching)?

In February, after more than a year consulting with a range of experts, the Department of Defense (DoD) released five principles for ethics around artificial intelligence (AI). If AI doesn't meet these standards, the Department has said, it won't be fielded. "The United States, together with our allies and partners, must accelerate the adoption of AI and lead in its national security applications to maintain our strategic position, prevail on future battlefields, and safeguard the rules-based international order," Secretary Mark Esper said in the news release. The principles, which apply to combat and non-combat functions, are that AI must be the following: responsible, equitable, traceable, reliable, and governable. Such guidelines are relatively high level, though, leaving individual departments and agencies on their own to implement what each adjective means for a specific use case.

In February, after more than a year consulting with a range of experts, the Department of Defense (DoD) released five principles for ethics around artificial intelligence (AI). If AI doesn't meet these standards, the Department has said, it won't be fielded. "The United States, together with our allies and partners, must accelerate the adoption of AI and lead in its national security applications to maintain our strategic position, prevail on future battlefields, and safeguard the rules-based international order," Secretary Mark Esper said in the news release. The principles, which apply to combat and non-combat functions, are that AI must be the following: responsible, equitable, traceable, reliable, and governable. Such guidelines are relatively high level, though, leaving individual departments and agencies on their own to implement what each adjective means for a specific use case.

The ethics of AI and robotics is often focused on "concerns" of various sorts, which is a typical response to new technologies. Many such concerns turn out to be rather quaint (trains are too fast for souls); some are predictably wrong when they suggest that the technology will fundamentally change humans (telephones will destroy personal communication, writing will destroy memory, video cassettes will make going out redundant); some are broadly correct but moderately relevant (digital technology will destroy industries that make photographic film, cassette tapes, or vinyl records); but some are broadly correct and deeply relevant (cars will kill children and fundamentally change the landscape). The task of an article such as this is to analyse the issues and to deflate the non-issues. Some technologies, like nuclear power, cars, or plastics, have caused ethical and political discussion and significant policy efforts to control the trajectory these technologies, usually only once some ...

Deep neural networks ("deep learning") have emerged as a technology of choice to tackle problems in natural language processing, computer vision, speech recognition and gameplay, and in just a few years has led to superhuman level performance and ushered in a new wave of "AI." Buoyed by these successes, researchers in the physical sciences have made steady progress in incorporating deep learning into their respective domains. However, such adoption brings substantial challenges that need to be recognized and confronted. Here, we discuss both opportunities and roadblocks to implementation of deep learning within materials science, focusing on the relationship between correlative nature of machine learning and causal hypothesis driven nature of physical sciences. We argue that deep learning and AI are now well positioned to revolutionize fields where causal links are known, as is the case for applications in theory. When confounding factors are frozen or change only weakly, this leaves open the pathway for effective deep learning solutions in experimental domains. Similarly, these methods offer a pathway towards understanding the physics of real-world systems, either via deriving reduced representations, deducing algorithmic complexity, or recovering generative physical models. However, extending deep learning and "AI" for models with unclear causal relationship can produce misleading and potentially incorrect results. Here, we argue the broad adoption of Bayesian methods incorporating prior knowledge, development of DL solutions with incorporated physical constraints, and ultimately adoption of causal models, offers a path forward for fundamental and applied research. Most notably, while these advances can change the way science is carried out in ways we cannot imagine, machine learning is not going to substitute science any time soon.

Despite the potential of Machine learning (ML) to learn the behavior of malware, detect novel malware samples, and significantly improve information security (InfoSec) we see few, if any, high-impact ML techniques in deployed systems, notwithstanding multiple reported successes in open literature. We hypothesize that the failure of ML in making high-impacts in InfoSec are rooted in a disconnect between the two communities as evidenced by a semantic gap---a difference in how executables are described (e.g. the data and features extracted from the data). Specifically, current datasets and representations used by ML are not suitable for learning the behaviors of an executable and differ significantly from those used by the InfoSec community. In this paper, we survey existing datasets used for classifying malware by ML algorithms and the features that are extracted from the data. We observe that: 1) the current set of extracted features are primarily syntactic, not behavioral, 2) datasets generally contain extreme exemplars producing a dataset in which it is easy to discriminate classes, and 3) the datasets provide significantly different representations of the data encountered in real-world systems. For ML to make more of an impact in the InfoSec community requires a change in the data (including the features and labels) that is used to bridge the current semantic gap. As a first step in enabling more behavioral analyses, we label existing malware datasets with behavioral features using open-source threat reports associated with malware families. This behavioral labeling alters the analysis from identifying intent (e.g. good vs bad) or malware family membership to an analysis of which behaviors are exhibited by an executable. We offer the annotations with the hope of inspiring future improvements in the data that will further bridge the semantic gap between the ML and InfoSec communities.

In recent times many state-of-the-art machine learning models have been shown to be fragile to adversarial attacks. In this work we attempt to build our theoretical understanding of adversarially robust learning with neural nets. We demonstrate a specific class of neural networks of finite size and a non-gradient stochastic algorithm which tries to recover the weights of the net generating the realizable true labels in the presence of an oracle doing a bounded amount of malicious additive distortion to the labels. We prove (nearly optimal) tradeoffs among the magnitude of the adversarial attack, the accuracy and the confidence achieved by the proposed algorithm. The seminal paper [35] was among the first to highlight a key vulnerability of state-of-the-art network architectures like GoogLeNet, that adding small imperceptible adversarial noise to test data can dramatically impact the performance of the network.

Accurate forecasts of COVID-19 is central to resource management and building strategies to deal with the epidemic. We propose a heterogeneous infection rate model with human mobility for epidemic modeling, a preliminary version of which we have successfully used during DARPA Grand Challenge 2014. By linearizing the model and using weighted least squares, our model is able to quickly adapt to changing trends and provide extremely accurate predictions of confirmed cases at the level of countries and states of the United States. We show that during the earlier part of the epidemic, using travel data increases the predictions. Training the model to forecast also enables learning characteristics of the epidemic. In particular, we show that changes in model parameters over time can help us quantify how well a state or a country has responded to the epidemic. The variations in parameters also allow us to forecast different scenarios such as what would happen if we were to disregard social distancing suggestions.