We live in the greatest time in human history. Only 200 years ago, for most Europeans, life was a struggle rather than a pleasure. Without antibiotics and hospitals, every infection was fatal. There was only a small elite of citizens who lived in the cities in relative prosperity. Freedom of opinion, human and civil rights were far away. Voting rights and decision-making were reserved for a class consisting of nobility, clergy, the military and rich citizens. The interests of the general population were virtually ignored.

AI and machine learning have developed significantly in recent years. In fact, so profound is the transformation that reports now claim that up to 1.5 million jobs are at risk from being replaced by automation. Industries around the world are set to be transformed by AI. The rise of legal artificial intelligence is just one such example. In most cases, the danger is exaggerated; outside of a few vulnerable industries, the focus will largely be on automating tasks within jobs, rather than the jobs themselves – at least in the near future. But to what extent will the legal industry be affected by automation?

Machines that make decisions about us: what could possibly go wrong? Essays, speeches, seminars pose that question year after year as artificial intelligence research makes stunning advances. Baked-in biases in algorithms are only one of many issues as a result. Jonathan Shaw, managing editor, Harvard Magazine, wrote earlier this year: "Artificial intelligence can aggregate and assess vast quantities of data that are sometimes beyond human capacity to analyze unaided, thereby enabling AI to make hiring recommendations, determine in seconds the creditworthiness of loan applicants, and predict the chances that criminals will re-offend." Again, what could possibly go wrong?

The biggest tech companies want you to know that they're taking special care to ensure that their use of artificial intelligence to sift through mountains of data, analyze faces or build virtual assistants doesn't spill over to the dark side. But their efforts to assuage concerns that their machines may be used for nefarious ends have not been universally embraced. Some skeptics see it as mere window dressing by corporations more interested in profit than what's in society's best interests. "Ethical AI" has become a new corporate buzz phrase, slapped on internal review committees, fancy job titles, research projects and philanthropic initiatives. The moves are meant to address concerns over racial and gender bias emerging in facial recognition and other AI systems, as well as address anxieties about job losses to the technology and its use by law enforcement and the military.

As AI systems become prevalent in high stakes domains such as surveillance and healthcare, researchers now examine how to design and implement them in a safe manner. However, the potential harms caused by systems to stakeholders in complex social contexts and how to address these remains unclear. In this paper, we explain the inherent normative uncertainty in debates about the safety of AI systems. We then address this as a problem of vagueness by examining its place in the design, training, and deployment stages of AI system development. We adopt Ruth Chang's theory of intuitive comparability to illustrate the dilemmas that manifest at each stage. We then discuss how stakeholders can navigate these dilemmas by incorporating distinct forms of dissent into the development pipeline, drawing on Elizabeth Anderson's work on the epistemic powers of democratic institutions. We outline a framework of sociotechnical commitments to formal, substantive and discursive challenges that address normative uncertainty across stakeholders, and propose the cultivation of related virtues by those responsible for development.

Mitigating prejudicial and abusive behavior online is no easy feat, given the level of toxicity in some communities. More than one in five respondents in a recent survey reported being subjected to physical threats, and nearly one in five experienced sexual harassment, stalking, or sustained harassment. Of those who experienced harassment, upwards of 20% said it was the result of their gender identity, race, ethnicity, sexual orientation, religion, occupation, or disability. In pursuit of a solution, Jigsaw -- the organization working under Google parent company Alphabet to tackle cyber bullying, censorship, disinformation, and other digital issues of the day -- today released what it claims is the largest public data set of comments and annotations with toxicity labels and identity labels. It's intended to help measure bias in AI comment classification systems, which Jigsaw and others have historically measured using synthetic data from template sentences.

Some areas of AI are further along in adoption than others. One of those areas is in recruiting. Already, there are companies that are marketing services to review hundreds (or thousands) of applicants and give each candidate a "score" based on multiple factors.The potential pitfall is that the output from some of these systems may have a disparate impact on a protected group. The most notable example was a system being developed (and rejected) by Amazon that did not like women. Thus, HR needs to have a seat at the table when these systems are being considered.

MELVILLE, N.Y., November 18, 2019 – Effectively managing today's workforce is crucial for improving customer experience, operational efficiency, and compliance. Yet currently, rising expectations of both customers and employees have made forecasting and scheduling contact center agents and customer engagement resources exponentially more challenging. To give companies a simpler way to manage work across the enterprise, Verint Systems Inc. (Nasdaq: VRNT), The Customer Engagement Company, today announced the newest release of its market-leading Workforce Management (WFM) solution, which leverages artificial intelligence-infused automation and new mobile tools to streamline forecasting and scheduling and improve employee engagement, all easily accessible via the Verint Cloud. "The workforce represents up to 80 percent of overall contact center budgets so accurate and cost-effective scheduling is vital," says Verint's John Goodson, SVP and general manager, Products. "At the same time, today's employees demand easier flex scheduling options, so organizations must balance flexibility and cost to provide superior service. As a pioneer in WFM, we view this new release as one that can invigorate the market to meet the ever-changing demands of today's contact centers and throughout the enterprise."

How often, or little, they're used by employees to increase their efficiency and improve customer satisfaction. For IBM, the only way to make AI a core part of the workflow across entire companies is to take a cue from a decades-old office staple: email. It's dangerous to think of AI as a magic tool, said Daniel Hernandez, VP of data and AI at IBM, speaking at the Gartner IT Symposium/Xpo in Orlando, Florida last week. Instead, it should be seen as a strategy to empower staffers, helping them make more effective decisions through data while boosting employee experience. "Email gets no respect because it's boring," said Hernandez.

Human biases are well-documented, from implicit association tests that demonstrate biases we may not even be aware of, to field experiments that demonstrate how much these biases can affect outcomes. Over the past few years, society has started to wrestle with just how much these human biases can make their way into artificial intelligence systems -- with harmful results. At a time when many companies are looking to deploy AI systems across their operations, being acutely aware of those risks and working to reduce them is an urgent priority. The problem is not entirely new. Back in 1988, the UK Commission for Racial Equality found a British medical school guilty of discrimination.