When making everyday decisions, people are guided by their conscience, an internal sense of right and wrong. By contrast, artificial agents are not currently endowed with a moral sense. As a consequence, they may unknowingly act immorally, especially when trained on environments that disregard moral concerns such as violent video games. With the advent of generally capable agents that pretrain on many environments, it will become necessary to mitigate inherited biases from such environments that teach immoral behavior. To facilitate the development of agents that avoid causing wanton harm, we introduce Jiminy Cricket, an environment suite of 25 text-based adventure games with thousands of diverse, morally salient scenarios. By annotating every possible game state, the Jiminy Cricket environments robustly evaluate whether agents can act morally while maximizing reward. Using models with commonsense moral knowledge, we create an elementary artificial conscience that assesses and guides agents. In extensive experiments, we find that the artificial conscience approach can steer agents towards moral behavior without sacrificing performance.

The rise of human-information systems, cybernetic systems, and increasingly autonomous systems requires the application of epistemic frameworks to machines and human-machine teams. This chapter discusses higher-order design principles to guide the design, evaluation, deployment, and iteration of Lethal Autonomous Weapons Systems (LAWS) based on epistemic models. Epistemology is the study of knowledge. Epistemic models consider the role of accuracy, likelihoods, beliefs, competencies, capabilities, context, and luck in the justification of actions and the attribution of knowledge. The aim is not to provide ethical justification for or against LAWS, but to illustrate how epistemological frameworks can be used in conjunction with moral apparatus to guide the design and deployment of future systems. The models discussed in this chapter aim to make Article 36 reviews of LAWS systematic, expedient, and evaluable. A Bayesian virtue epistemology is proposed to enable justified actions under uncertainty that meet the requirements of the Laws of Armed Conflict and International Humanitarian Law. Epistemic concepts can provide some of the apparatus to meet explainability and transparency requirements in the development, evaluation, deployment, and review of ethical AI.

Much recent research has uncovered and discussed serious concerns of bias in facial analysis technologies, finding performance disparities between groups of people based on perceived gender, skin type, lighting condition, etc. These audits are immensely important and successful at measuring algorithmic bias but have two major challenges: the audits (1) use facial recognition datasets which lack quality metadata, like LFW and CelebA, and (2) do not compare their observed algorithmic bias to the biases of their human alternatives. In this paper, we release improvements to the LFW and CelebA datasets which will enable future researchers to obtain measurements of algorithmic bias that are not tainted by major flaws in the dataset (e.g. identical images appearing in both the gallery and test set). We also use these new data to develop a series of challenging facial identification and verification questions that we administered to various algorithms and a large, balanced sample of human reviewers. We find that both computer models and human survey participants perform significantly better at the verification task, generally obtain lower accuracy rates on dark-skinned or female subjects for both tasks, and obtain higher accuracy rates when their demographics match that of the question. Computer models are observed to achieve a higher level of accuracy than the survey participants on both tasks and exhibit bias to similar degrees as the human survey participants.

Model efficiency is a critical aspect of developing and deploying machine learning models. Inference time and latency directly affect the user experience, and some applications have hard requirements. In addition to inference costs, model training also have direct financial and environmental impacts. Although there are numerous well-established metrics (cost indicators) for measuring model efficiency, researchers and practitioners often assume that these metrics are correlated with each other and report only few of them. In this paper, we thoroughly discuss common cost indicators, their advantages and disadvantages, and how they can contradict each other. We demonstrate how incomplete reporting of cost indicators can lead to partial conclusions and a blurred or incomplete picture of the practical considerations of different models. We further present suggestions to improve reporting of efficiency metrics.

Machine learning models have been criticized for reflecting unfair biases in the training data. Instead of solving for this by introducing fair learning algorithms directly, we focus on generating fair synthetic data, such that any downstream learner is fair. Generating fair synthetic data from unfair data-- while remaining truthful to the underlying data-generating process (DGP) --is non-trivial. In this paper, we introduce DECAF: a GAN-based fair synthetic data generator for tabular data. With DECAF we embed the DGP explicitly as a structural causal model in the input layers of the generator, allowing each variable to be reconstructed conditioned on its causal parents. This procedure enables inference-time debiasing, where biased edges can be strategically removed for satisfying user-defined fairness requirements. The DECAF framework is versatile and compatible with several popular definitions of fairness. In our experiments, we show that DECAF successfully removes undesired bias and-- in contrast to existing methods --is capable of generating high-quality synthetic data. Furthermore, we provide theoretical guarantees on the generator's convergence and the fairness of downstream models.

Nearly two-thirds of Americans want the U.S to regulate the development and use of artificial intelligence in the next year or sooner -- with half saying that regulation should have begun yesterday, according to a Morning Consult poll. Another 13% say that regulation should start in the next year. "You can thread this together," Austin Carson, founder of new nonprofit group SeedAI and former government relations lead for Nvidia, said in an email. "Half or more Americans want to address all of these things, split pretty evenly along ideological lines." The poll, which SeedAI commissioned, backs up earlier findings that while U.S. adults support investment in the development of AI, they want clear rules around that development.

Every time a dramatic, unforeseen political event happens, there follows a left-field fixation that some out-of-control technology created it. Whenever this fear about big tech comes around we are told that something new, even more toxic, has infiltrated our public discourse, triggering hatred towards politicians and public figures, conspiracy theories about Covid and even major political events like Brexit. The concern over anonymity online becomes a particular worry – as if ending it will somehow, like throwing a blanket at a raging house fire, subdue our fevered state. You may remember that during the summer's onslaught of racist abuse towards black players in the England football team, instead of reckoning with the fact that racism still haunts this country, we busied ourselves with bluster about how "cowards" online would be silenced if we only just demanded they identify themselves. We resort to this explanation, that shadowy social media somehow stimulate our worst impulses, despite there being little evidence that most abuse is from unidentifiable sources.

A recent study in Nature Machine Intelligence by researchers at Carnegie Mellon sought to investigate the impact that mitigating bias in machine learning has on accuracy. Despite what researchers referred to as a "commonly held assumption" that reducing disparities requires either accepting a drop in accuracy or developing new, complex methods, they found that the trade-offs between fairness and effectiveness can be "negligible in practice." "You actually can get both. You don't have to sacrifice accuracy to build systems that are fair and equitable," said Rayid Ghani, a CMU computer science professor and an author on the study, in a statement. At the same time, Ghani noted, "It does require you to deliberately design systems to be fair and equitable.

In this study, we propose using a neural embedding model-graph neural network (GNN)- that leverages the heterogeneous features of urban areas and their interactions captured by human mobility network to obtain vector representations of these areas. Using large-scale high-resolution mobility data sets from millions of aggregated and anonymized mobile phone users in 16 metropolitan counties in the United States, we demonstrate that our embeddings encode complex relationships among features related to urban components (such as distribution of facilities) and population attributes and activities. The spatial gradient in each direction from city center to suburbs is measured using clustered representations and the shared characteristics among urban areas in the same cluster. Furthermore, we show that embeddings generated by a model trained on a different county can capture 50% to 60% of the emergent spatial structure in another county, allowing us to make cross-county comparisons in a quantitative way. Our GNN-based framework overcomes the limitations of previous methods used for examining spatial structures and is highly scalable. The findings reveal non-linear relationships among urban components and anisotropic spatial gradients in cities. Since the identified spatial structures and gradients capture the combined effects of various mechanisms, such as segregation, disparate facility distribution, and human mobility, the findings could help identify the limitations of the current city structure to inform planning decisions and policies. Also, the model and findings set the stage for a variety of research in urban planning, engineering and social science through integrated understanding of how the complex interactions between urban components and population activities and attributes shape the spatial structures in cities.

Parliament is working on the Commission proposal, presented on 21 April 2021, for turning Europe into the global hub for trustworthy AI. Ahead of the Commission's proposal on AI, the Parliament set up a special committee to analyse the impact of artificial intelligence on the EU economy. "Europe needs to develop AI that is trustworthy, eliminates biases and discrimination, and serves the common good, while ensuring business and industry thrive and generate economic prosperity," said the new committee chair Dragoș Tudorache. On 20 October 2020, Parliament adopted three reports outlining how the EU can best regulate AI while boosting innovation, ethical standards and trust in technology. One of the reports focuses on how to ensure safety, transparency and accountability, prevent bias and discrimination, foster social and environmental responsibility, and ensure respect for fundamental rights.