Daily Tech Digest - September 13, 2020

The Importance Of Predictive AI In Cybersecurity

The utilization of AI systems, in the realm of cybersecurity, can have three kinds of impact, it is constantly expressed in the work: «AI can: grow cyber threats (amount); change the run of the mill character of these dangers (quality); and present new and obscure dangers (quantity and quality). Artificial intelligence could grow the set of entertainers that are fit for performing noxious cyber activities, the speed at which these actors can play out the exercises, and the set of plausible targets. Fundamentally, AI-fueled cyber attacks could likewise be available in more powerful, finely targeted and advanced activities because of the effectiveness, scalability and adaptability of these solutions. Potential targets are all the more effectively identifiable and controllable. In a mix of defensive techniques and cyber threat detection, AI will move towards predictive techniques that can identify Intrusion Detection Systems (IDS) pointed toward recognizing illegal activity within a computer or network, or spam or phishing with two-factor authentication systems. The guarded strategic utilization of AI will likewise focus soon on automated vulnerability testing, also known as fuzzing.

Top 9 ways RPA and analytics work together

Increasingly, organizations are using robotic process automation in analytics tasks from assembling data spread across the company to analyzing how business processes work and how they can improve. "RPA is helping streamline the processes that create valuable insights, changing what areas analytics are measuring and helping to find new domains of time-consuming tasks to focus on," said Michael Shepherd, engineer at Dell Technologies Services. When it comes to RPA and analytics, the automation tool should be a complement to, rather than a replacement for, an integrated data platform across the company, said Jonathan Hassell, content director for data and AI at O'Reilly Media. When companies lack an integrated data platform, it makes analytics more difficult overall. "RPA can help in some ways, but the potential for RPA to unlock insights in data and further output and processes requires a good data platform with great health and hygiene," he said. Hassell recommended organizations look at three key ways RPA can change analytics. First, it helps create better data from the outset. Second, the organization can deploy it in the context of machine learning to sift through large quantities of data and identify useful information for humans to look at.

Rethinking AI Ethics - Asimov has a lot to answer for

AI per se is neither ethical or not, it is how it is applied by practitioners. But if an AI model is introduced into the market that routinely disrupts ethical concepts, what is it? A neutrally ethical artifact produced by unethical people? Think about the judicial system COMPAS that routinely made bail and sentencing recommendations two- to three-times more severe for African-Americans. That is clearly an unethical AI application. Instead of teaching people about the ethics of Plato, and Aristotle and Kant and trying to draw a line from that to build AI applications, a better approach is to start with identifying developers who are simply good people. People who show forbearance, who have the backbone to resist their organization's directive to deliver wrong things. People who have prudence can look beyond the results of a model and project how it will affect the world. Only good people should develop AI. A quick search turns up over a hundred AI Ethics proclamations from governments, non-profit special interest groups, government committees, etc, and it's consuming too much energy and bandwidth. In fact, the cynical view is that all of this is just a way to avoid authorities from issuing rules for AI.

Designing for Privacy — An Emerging Software Pattern

The first thing to do is to separate, and consider differently, the user data that your system needs to “know” from the user data that the system can collect and treat without actually having access to it. We call those two categories: Known and Unknown User Data. There is no magic recipe for this separation as it depends on your system’s functionality. There are systems where all user data can be considered Unknown — where the system has no knowledge of the user to whom it delivers its functionality. Yet, most systems need to identify the user in order to make him pay for a service that they deliver. A ride sharing platform might want to consider the identity of riders and drivers “known” but the destination of the ride and any messages exchanged between drivers and riders “unknown”. A hotel booking platform might perform the split in such a way to connect users with hotels, get a fee from hotels, but ignore the dates of the booking that reveal the user’s whereabouts. Once you segment which user data to treat as “known” and which as “unknown” you can adopt a new flavor of client-server architecture — the one when you treat the “known” data as you normally would, but where the “unknown” user data is kept at the user’s endpoint; 

Blockchain may break EU privacy law—and it could get messy

This technology’s transparency and immutability, meaning it cannot be edited later, is one of its biggest selling points. It’s also why massive enterprises have been drawn to using it. But if someone wanted to invoke their right to be forgotten—and order entries about themselves to be erased from the blockchain—networks may be duty bound to comply under court order. Here’s the kicker: in some cases, it may be near impossible to obey these orders because of the sheer levels of computing power required to edit a blockchain. And in others, the decentralized nature of some networks may mean it’s impossible to pin down someone who can be held responsible for fulfilling the court order. As part of her research, Dr Wahlstrom looked at a variation of blockchain technology that is known as Holochain. She concluded that it could be more compatible with the “right to be forgotten” legislation because of how its distributed database breaks the blockchain up—meaning it is easier for a smaller node to prevent contested data from being reshared. “This allows individuals to verify data without disclosing all its details or permanently storing it in the cloud,” she added.

RBI seeks exemption from data protection law

The banking regulator has also gone a step further and suggested that instead of the Central government, “sectoral regulators be given the power to classify personal data as critical.” Any critical data, according to the proposed act, can be processed only in India. Objecting to classification of financial data as sensitive personal data , RBI’s note maintained that this would lead to higher compliance and explicit consent, which “would translate to increase in costs for providing services to customers. Financial inclusion efforts rely on lower service charges for offering basic banking services. The increase in costs would compel banks to increase the charges associated with offering banking services.” RBI’s note also pointed out that countries such as the UK, France, Germany and Italy do not make such a classification. Privacy experts said the RBI cannot legally claim an exemption from the obligations that stem from the 2017 Supreme Court ruling as the Puttuswamy judgment, which upheld privacy as a fundamental right for Indian citizens. “What the bill does is flesh out that right in terms of the actual actions that need to be done. 

Digital transformation takes more than the wave of a wand

Championing digital transformation requires more than a magic wand or “plug and play” approach. Even the best technology is virtually worthless if everyone isn’t able, available and on board to use it. Without the proper talent, employee training and company-wide desire to evolve in place, people will inevitably revert to their old ways, using antiquated and siloed tools like Excel. Plus, the systems you already have in place have to keep running smoothly as you roll out digitization plans. Digital transformation does not happen with the flip of a switch. It requires ongoing strategic efforts to create a balance among new technologies, strategic solutions and traditional systems. This is why the aforementioned culture shift is essential before starting your transformation journey. As Raconteur author Ben Rossi says in this “Digital innovation and the supply chain” report, “A top-down mandate from board level to drive supply chain transformation is critical to getting the rest of the company to collaborate and change their mindset. Without that, heads of supply chain will run into resistance to change, in turn reducing the chance of achieving the broader transformation goals.” 

Genetic Algorithms: a developer perspective

There are various interesting theories regarding convergence in evolutionary algorithms, but these are of no concern to us here. Our interest is in understanding how these algorithms may be used to solve Artificial Intelligence problems, rather than in understanding why they actually work. One important class of evolutionary algorithms used in practical applications is genetic algorithms: these stress the importance of the data representation used to encode possible solutions to our optimization problem. The class name is inspired by an analogy with genetic code – the material that encodes our ‘phenotype’ or physical appearance. The use of the adjective ‘genetic’ reflects the fact that evolving solutions are represented by data structures, usually strings, reminiscent of biological genetic code. ... The goal of a genetic algorithm is to discover a phenotype that maximises fitness, by allowing a certain population to evolve across several generations. The next question is: how does the evolution of individuals happen? Genetic algorithms apply a set of ‘genetic operations’ to chromosomes of each generation to allow them to reproduce and, in the process, introduce casual mutations, much as occurs in most living beings.

The Story of Data — Privacy By Design

Every byte of data has a story to tell. The question is whether the story is being narrated accurately and securely. Usually, we focus sharply on the trends around data with a goal of revenue acceleration but commonly forget about the vulnerabilities caused due to bad data management. Data possesses immense power, but immense power comes with increased responsibility. In today’s world collecting, analyzing and build prediction models is simply not enough. I keep reminding my students that we are in a generation where the requirements for data security have perhaps surpassed the need for data correctness. Hence the need for Privacy By Design is greater than ever. ... Until recently businesses have focused on looking at data over long stretches of time, made possible by Big Data. With the advent of Internet of Things (IoT) analyzing real-time data has gained immense importance. It is very common these days to have devices in our homes that collect personal data and transmit it to external locations for either monitoring or analytical purposes. In many cases the the poor consumer is finding it difficult to balance the benefits they get from surrendering their personal data against the risk involved with providing them.

Enterprise Data Literacy: Understanding Data Management

Sandwell believes that the Data Literacy problem stems from specialized information needs and a lack of shared context. He remarked: “Data Literacy affects all organizational levels. Everyone uses data for different reasons, including senior managers and the Chief Data Officer (CDO). The CDO tends to come from the business side and takes that perspective. However, he or she may have a steep learning curve about making technical infrastructure ready to serve and deliver.” On the technical side, workers have a good data inventory; however, they have less of an understanding of what the data contents mean to the business. Meanwhile the more data literate data scientists and business analysts put business and technical information together faster, with more direct data querying and manipulation. So, across the enterprise, everyone has a different Data Literacy perspective and talks at cross purposes to one other. Add to the situation various data maturity levels across departments and enterprises. Some ask about “the data on-hand, where to access it, and how it gets used and by whom.” Others have figured out these basics and have different questions on how to do Metadata Management and create a data catalog of all the data sets. Since everyone has different data requirements at different times, getting to a uniform Enterprise Data Literacy remains elusive.

Quote for the day:

"Leading people is like cooking. Don_t stir too much; It annoys the ingredients_and spoils the food." -- Rick Julian

No comments:

Post a Comment