Text IQ, a machine learning system that parses and understands sensitive corporate data, has raised $12.6 million in Series A funding led by FirstMark Capital, with participation from Sierra Ventures. Text IQ started as co-founder Apoorv Agarwal's Columbia thesis project titled "Social Network Extraction From Text." The algorithm he built was able to read a novel, like Jane Austen's "Emma," for example, and understand the social hierarchy and interactions between characters. This people-centric approach to parsing unstructured data eventually became the kernel of Text IQ, which helps corporations find what they're looking for in a sea of unstructured, and highly sensitive, data. The platform started as a tool used by corporate legal teams.

We already knew an artificial intelligence could reflect the racial bias of its creator. But San Francisco thinks the tech could potentially do the opposite as well, by identifying and counteracting racial prejudice -- and it plans to put the theory to the test in a way that could change the legal system forever. On Wednesday, San Francisco District Attorney George Gascon announced that city prosecutors will begin using an AI-powered "bias-mitigation tool" created by Stanford University researchers on July 1. This could include their last name, eye color, hair color, or location. It also removes any information that might identify the law enforcement involved in the case, such as their badge number, a DA spokesperson told The Verge. Prosecutors will look at these redacted reports, record their decision on whether to charge a suspect, and then see the unredacted report before making their final charging decision.

The overarching theme of'Nederland Digitaal Dag' was Artificial Intelligence (AI). As a Dutch AI company operating on a global level, Dashmote was invited to participate in the ongoing debate regarding this topic. Following the many sessions that took place, we would like to share four main takeaways from these sessions and our thoughts on them. Yes, you read that correctly. After the official kickoff, Google's Chief Economist Hal Varian took to the stage.

The exciting promise of personalization may not be here yet (at least not at scale), but it's not far off. Advances in technology, data, and analytics will soon allow marketers to create much more personal and "human" experiences across moments, channels, and buying stages. Physical spaces will be reconceived, and customer journeys will be supported far beyond a brand's front door. While these opportunities are exciting, most marketers feel underequipped to deliver. A recent McKinsey survey of senior marketing leaders finds that only 15 percent of CMOs believe their company is on the right track with personalization.

Lin Classon, Director of Public Cloud Product at managed cloud provider Ensono and former Googler, has spent her entire career attending technology conferences – places where, unlike the public restrooms at most events, women tend to have the toilet all to themselves. Only about 25 percent of the speakers and audience of the average tech conference are female. That the science, technology, engineering, and mathematics fields have a problem with discrimination, harassment, and inequality towards women is a well-documented and almost universally known fact. But in the field of AI, where a lack of representation directly leads to the development of inherently biased algorithms and systems, the effects are exacerbated. For the sake of all humans, we need to start being better allies for females working in machine learning -- and in general.

Existing graph-based methods for extractive document summarization represent sentences of a corpus as the nodes of a graph or a hypergraph in which edges depict relationships of lexical similarity between sentences. Such approaches fail to capture semantic similarities between sentences when they express a similar information but have few words in common and are thus lexically dissimilar. To overcome this issue, we propose to extract semantic similarities based on topical representations of sentences. Inspired by the Hierarchical Dirichlet Process, we propose a probabilistic topic model in order to infer topic distributions of sentences. As each topic defines a semantic connection among a group of sentences with a certain degree of membership for each sentence, we propose a fuzzy hypergraph model in which nodes are sentences and fuzzy hyperedges are topics. To produce an informative summary, we extract a set of sentences from the corpus by simultaneously maximizing their relevance to a user-defined query, their centrality in the fuzzy hypergraph and their coverage of topics present in the corpus. We formulate a polynomial time algorithm building on the theory of submodular functions to solve the associated optimization problem. A thorough comparative analysis with other graph-based summarization systems is included in the paper. Our obtained results show the superiority of our method in terms of content coverage of the summaries.

How is your firm currently using AI? The main AI tool we're currently using is Luminance. The solution was originally created to help with contract reviews, particularly in relation to due diligence exercises during mergers and acquisitions (M&A). But, because Luminance has a host of additional functionalities – and because you can train it– we're now using it in lots of different ways around Slaughter and May, including within the firm's support functions to optimize a variety of business processes. The use of predictive analytics in litigation – effectively predicting a dispute's outcome - is a technology that is at a very early stage of development, but it's something we're exploring.

A central concern surrounding AI is how it might affect the labour market. In recent years, technology that relies on automation has become more advanced, and its application is increasing across a range of different business settings. Does it pose a threat to business and what are the wider implications for society? AI has wide-ranging implications and not just in the places you might first expect. However, it's not necessarily true that AI will destabilize whole areas of the work force as it pertains to certain kinds of workers in the economy.

We have prepared this AI toolkit in collaboration with our colleagues at Linklaters. It is based on our shared experience of advising clients on these issues and deploying AI tools in our own business. It draws upon the expertise of lawyers from our technology, privacy, intellectual property, competition, employment and financial services regulatory groups. The toolkit uses Australian law as its reference point but draws on experiences from the EU and the insights apply equally in other jurisdictions. However, it does not consider autonomous vehicles or robotics, which raise their own regulatory and commercial issues. The toolkit starts with a short technical primer.

All financial industry analysts agree that the number of transactions paid using credit and debit cards will continue to grow annually. Moreover, this trend is bound to continue given the planned introduction of the new Apple Card, which is expected to become an instant hit with Millennials, especially those who currently do not have a credit card. Banks that process credit and debit card transactions, therefore, constantly must be on the lookout for new technologies that can handle all this new data traffic quickly and dependably. By every indication, they've found it. AI is quickly becoming central to emerging bank technologies.