Automatic reasoning about textual information is a challenging task in modern Natural Language Processing (NLP) systems. In this work we describe our proposal for representing and reasoning about Portuguese documents by means of Linked Data like ontologies and thesauri. Our approach resorts to a specialized pipeline of natural language processing (part-of-speech tagger, named entity recognition, semantic role labeling) to populate an ontology for the domain of criminal investigations. The provided architecture and ontology are language independent. Although some of the NLP modules are language dependent, they can be built using adequate AI methodologies.

Classifiers that achieve demographic balance by explicitly using protected attributes such as race or gender are often politically or culturally controversial due to their lack of individual fairness, i.e. individuals with similar qualifications will receive different outcomes. Individually and group fair decision criteria can produce counter-intuitive results, e.g. that the optimal constrained boundary may reject intuitively better candidates due to demographic imbalance in similar candidates. Both approaches can be seen as introducing individual resentment, where some individuals would have received a better outcome if they either belonged to a different demographic class and had the same qualifications, or if they remained in the same class but had objectively worse qualifications (e.g. lower test scores). We show that both forms of resentment can be avoided by using monotonically constrained machine learning models to create individually fair, demographically balanced classifiers.

In critical decision-making scenarios, optimizing accuracy can lead to a biased classifier, hence past work recommends enforcing group-based fairness metrics in addition to maximizing accuracy. However, doing so exposes the classifier to another kind of bias called infra-marginality. This refers to individual-level bias where some individuals/subgroups can be worse off than under simply optimizing for accuracy. For instance, a classifier implementing race-based parity may significantly disadvantage women of the advantaged race. To quantify this bias, we propose a general notion of $\eta$-infra-marginality that can be used to evaluate the extent of this bias. We prove theoretically that, unlike other fairness metrics, infra-marginality does not have a trade-off with accuracy: high accuracy directly leads to low infra-marginality. This observation is confirmed through empirical analysis on multiple simulated and real-world datasets. Further, we find that maximizing group fairness often increases infra-marginality, suggesting the consideration of both group-level fairness and individual-level infra-marginality. However, measuring infra-marginality requires knowledge of the true distribution of individual-level outcomes correctly and explicitly. We propose a practical method to measure infra-marginality, and a simple algorithm to maximize group-wise accuracy and avoid infra-marginality.

It's one of our most distinctive features, but it seems that your voice isn't safe from cybercriminals, if a recent case is anything to go by. In the case, cybercriminals developed an AI that mimicked a CEO's voice so well, that it was able to scam an energy firm out of hundreds of thousands of pounds. The Wall Street Journal reported the scam, which happened back in March, and saw criminals swindle a staggering $243,000 (£201,000). The fraudsters used AI to mimic a chief executive from the German parent company of an unnamed UK energy firm. This voice was so believable that the UK-based CEO was tricked into making a large transfer of money to the chief executive, via a Hungarian supplier.

Facial recognition technology will not be deployed at the King's Cross development in the future, following a backlash prompted by the site owner's admission last month that the software had been used in its CCTV systems. The developer behind the prestigious central London site said the surveillance software had been used between May 2016 and March 2018 in two cameras on a busy pedestrian street running through its heart. It said it had abandoned plans for a wider deployment across the 67-acre, 50-building site and had "no plans to reintroduce any form of facial recognition technology at the King's Cross Estate". The site became embroiled in the debate about the ethics of facial recognition three weeks ago after releasing a short statement saying its cameras "use a number of detection and tracking methods, including facial recognition". That made it one of the first landowners to acknowledge it was deploying the software, described by human rights groups as authoritarian, partly because it captures and analyses images of people without their consent.

There are more than 1.9 billion users logged in to YouTube every single month who watch over a billion hours of video every day. With this number of users, activity, and content, it makes sense for YouTube to take advantage of the power of artificial intelligence (AI) to help operations. Here are a few ways YouTube, owned by Google, uses artificial intelligence today. In the first quarter of this year, 8.3 million videos were removed from YouTube, and 76% were automatically identified and flagged by artificial intelligence classifiers. More than 70% of these were identified before there were any views by users.

Fox News Flash top headlines for September 1 are here. Check out what's clicking on Foxnews.com As Americans celebrate Labor Day 2019, robots are stealing their jobs, as are immigrants, as are cheap imports from China. The first puzzle is: if all of these nefarious forces of free markets are stealing jobs, how is it that there are more Americans employed than ever before? Today, there are over 151 million workers on U.S. payrolls.

John Borthwick (above) believes the convenience of today's smart assistants comes at a price far higher than the cost paid for the devices. 'It's hard to call it anything but surveillance,' says the former Time Warner exec Tech investor John Borthwick believes the convenience of today's smart assistants from Amazon, Google and Apple comes at a price far higher than the cost paid for the devices. 'From a consumer standpoint, user standpoint, is that these, these devices are being used for what's -- it's hard to call it anything but surveillance,' Borthwick says, warning that government regulation may be the only safeguard to user privacy. Borthwick, a venture capitalist who started out in the technology industry with a web content studio that was bought by AOL, and who later headed tech strategy for Time Warner, tells Yahoo that he expects regulators will hand over more control of privacy to device users. As it stands now, he warns tech companies that manufacture and sell popular smart speakers, like Amazon's Echo, Google Assistant and Apple's HomePod, are having much more than they're audible responses recorded.

The U.S. Department of Commerce's Patent and Trademark Office (USPTO) is asking for the help of experts and the broader public to determine the impact AI will have on intellectual property and "whether new forms of intellectual property protection are needed." A call for public comment was published in the Federal Registrar by the USPTO today in search of answers about such issues as how AI is reshaping perceptions of inventions or whether additional information should be required to claim a deep learning system as an invention since they can have a large number of hidden layers and weights that evolve. To help solicit responses, the notice in the federal registrar comes along with a series of questions such as "what is an AI invention and what does it contain?" "What are the different ways that a natural person can contribute to conception of an AI invention and be eligible to be a named inventor? Structuring data in order to train a model?

Science fiction has long imagined a future in which humans constantly interact with robots and intelligent machines. This future is already happening in warehouses and manufacturing businesses. Other workers use virtual or augmented reality as part of their employment training, to assist them in performing their job or to interact with clients. And lots of workers are under automated surveillance from their employers. All that automation yields data that can be used to analyze workers' performance.