Will Privacy Regulations Stifle AI Innovation?

AI relies on quality data – and the more of it, the better. But that can be easier said than done now that recent privacy regulations like General Data Protection Regulation (GDPR) are requiring organizations to mask personal data. GDPR is a European regulation developed to protect privacy and make data collection and management more transparent and secure.

While GDPR is a European regulation, the U.S. may well be following suit. Already, several U.S. states have recently introduced legislation to expand data privacy rules and more states are expected to join over the coming months.

During a European Union privacy conference, Apple’s Tim Cooke just recently issued a call to action for U.S.-wide data-protection regulation, saying individuals’ personal information has been “weaponized.” According to the same article, U.S. Senator Mark Warner said he was encouraged by Microsoft, Apple and others’ support of regulation. He said, “Too often we’ve heard companies whose business models depend on intrusive and opaque collection of user data claim that any changes to the status quo will radically undermine American innovation, but Apple and others demonstrate that innovation doesn’t have to be a race to the bottom when it comes to data protection and user rights.”

But how will GDPR really impact AI? What happens, for example, when specific information needs to be collected on individuals to predict customer behavior, such as who might be likely to purchase a product or upgrade technology? Besides allowing people to request that companies remove their data, GDPR also requires companies to anonymize their data, unless identifying information is crucial to its worthiness.

This is especially true when it comes to AI in healthcare. According to an article, the GDPR introduced a right to explanation, which means that the logic of an automated decision can be challenged and the results tested – so businesses will need to think carefully before building an AI solution that cannot explain itself. Where required by GDPR, privacy impact assessments will be needed and privacy will take on even more urgency.

To conform to data privacy needs, professionals in all industries working with big data need to take out identifying details before processing the information. Similarly, the businesses using them should ensure that there is training or verify their workers’ knowledge in handling big data to avoid ethical violations and significant fines.

Maybe It’s not all bad news for AI innovation

There is a lot of speculation out there about what privacy regulations will mean for AI, but ironically, the regulations that some feel are stifling innovation may actually be forcing businesses to get their data houses in order. Consider that one of the main barriers to AI is data collection and many companies don’t know what type of data they have, let alone where it is or in what shape it is in. It’s common for different departments or business units to have their own data silos, leaving everyone to guess about what information is available in their organization.

Now, GDPR and other regulations are forcing companies to take a long hard look at their data and get their houses in order. They have to think critically about their data, what type they are storing, and what rules they are implementing to protect it. They need to find out where the data is, what it contains, how it is being used and ensure that the quality is good. Accomplishing all of these things and unlocking the data that has been stuck inside an organization is critical for any AI effort.

There’s always another way

The market is finding a way to unlock all of this hidden data – and to do so securely. Technology like synthetic data generation lets companies access personal information without identifying any individuals. Apple uses differential privacy to gather information on a group of users without using individual information, and Google offers a data loss prevention technology that strips personal information from databases.

Vendors are continuing to provide incentives for individuals to share their data. While that’s nothing new – it’s been around since loyalty cards were introduced – they are finding new and more attractive ways to encourage customers to opt in with their information.

Retail and e-commerce industries, in particular, will have a hard time adjusting to new privacy regulations because it is not anything they have ever had to do before. They can learn from industries like healthcare and financial services that have had to grapple with these data privacy issues for a while and figured out how to manage data privacy effectively and still mine the data they need. For example, healthcare organizations know how to share information in files while masking personal data.

How personal do you want to get?

When it comes down to it, how often do you really need to know about a specific customer? To determine what products to offer and the order in which to display website content, for example, companies need to know about patterns of behavior, and in some cases where individuals are located, but very specific and personalized data may not necessarily be needed. After all, data is needed to train algorithms, not to expose specific individuals. While AI innovation is being fueled to new heights thanks to solid, quality data, individual privacy need not be sacrificed in the process.

Why It Pays to Outsource Software Development Close to Home

While the boundaries to global commerce are clearly eroding, and no longer do companies operate only within the confines of their own countries, the fact of the matter remains that all is not as seamless as it may seem.

Changing global regulations about data sharing, complex AI-based technologies, a shortage of skilled data scientists and the need for greater collaboration between companies and their service providers are driving a renewed focus on U.S.-based nearshoring as the optimum way to reach business goals.

Companies are turning to U.S.-based nearshore providers because despite the growth in traditional offshoring, there have been bumps in the road. In some areas where the cost of labor is cheap, the workforce might be less educated and skilled, and they may rely on outdated frameworks and technologies, resulting in lower quality standards. And, with the rapid pace of business, distance can be a problem.

So how exactly are new market factors driving a resurgence in U.S.-based nearshoring?

Data privacy and other regulatory controls.

The growth of data-driven businesses and the ability to collect data on virtually anyone, has resulted in stricter data privacy regulations everywhere and the U.S. is no exception. For example, the U.S. telecommunications industry has rules governing the ability to share data overseas, and that includes sending data to software development providers. Since most of the AI applications that are created today, especially machine learning and analytics tools, require tons of training data it can be a major problem when the data can’t be easily sent to the provider without massive cleansing, obfuscation and encryption. Nearshoring to U.S. providers can eliminate this problem, enabling seamless transfer of vital data that fuels AI-based applications.

Complex AI-based technologies.

Given the rapid pace of technology advances, companies are under pressure to continually enhance software to improve business value, provide market differentiation and competitive advantage. Yet, emerging solutions that incorporate machine learning, chatbots and other smart apps are complicated to build and maintain, requiring the specialized expertise of software engineers and data scientists – who must remain involved for long-term training of AI algorithms. Because of the expertise that’s required for these cognitive solutions, as well as the need to truly understand specific business needs, working with nearshore providers that can relate to the local challenges and help companies derive valuable insights from smart solutions is becoming more crucial than ever.

A shortage of data scientists.

LinkedIn calculates that, in August 2018, employers were seeking 151,717 more data scientists than exist in the U.S. And, according to an article on The Quant Crunch report, demand is expected to rise 28% by 2020. These statistics indicate a major talent shortage when it comes to filling data science positions. Yet even without a talent shortage, many companies don’t find it cost-effective to hire internal data scientists to constantly maintain and grow AI solutions. Both of these factors are creating new demand for nearshore providers that have both the skill-set and understanding of local business needs to drive continued evolution of AI apps.

The need for collaboration.

The development of complex AI-based apps often requires close communication between the customer and offshore teams. While a major reason for the failure of software projects is a breakdown in communication, it’s hard to have real-time communication with an outsourcing partner when they are located several time zones away. Companies that conduct nearshoring from the U.S. typically experience less integration, cultural differences and other risks than European and APAC companies that more frequently outsource to neighboring countries with significantly different languages, currencies and regulatory requirements.


Nearshoring is clearly on the rise when it comes to today’s business needs for advanced software development. But to help ensure that they have a positive engagement, companies should consider the following in their software development outsourcing strategy:

  • The education and skill of the workforce – with emphasis on quality commitments
  • Specialized expertise and resources – especially when it comes to machine learning, deep learning and other specialties
  • Language and cultural barriers
  • Alignment between your approaches
  • Proximity and the need for ongoing communication
  • Regulatory issues. For example, in the U.S. government mandates require that work for many aerospace and defense contractors and healthcare providers is conducted by U.S. citizens, which would include U.S. territories such as Puerto Rico and the U.S. Virgin Islands.

With today’s fast pace of technology innovation, driven by AI solutions that need to think like the people they are supporting, companies know they can’t do it alone and are realizing the benefits of outsourcing in its many forms. Yet, we can expect to see the nearshoring segment of the market take off at lightning speed as companies begin to more fully grasp the added value of having a software partner closer to home, abiding by the same regulations and business practices and ultimately becoming a real extension of the business, as well as contributor to its overall success.

Chatbots Get an A+ for Increasing Enrollment, Student Satisfaction in Higher Ed

Given declining enrollment in higher ed institutions nationwide, universities are under increasing pressure to attract students and keep them happy. Customer service has become more important than ever, but with limited budgets and staff, many universities have fallen behind and need to up their game.

Let’s face it, millennials, who grew up with the Internet and self-service applications, are used to getting information how and when they want it, 24/7. They want to be able to easily find answers to their questions on academic requirements, financial aid and other areas at the tap of a finger on their mobile devices.

Universities have tried to staff call centers as best they can to address these questions, often using students whose first priority, of course, is their academic work. It’s just simply not feasible –not to mention exorbitantly expensive — to staff these call centers 24/7, so throwing more people at the problem isn’t the solution. And often, students don’t want to talk anyway. They want to use their mobile devices to get answers to their questions.

Fortunately, chatbots are an ideal way to address the problem. Chatbots, which simulate human communication either via voice or text, provide a direct user experience without any intermediaries. They can be programmed to answer all types of questions that students may have, providing immediate information to address their needs – through text and SMS messages, chat discussions on the website or by phone.

Making the best first impression for recruitment

Each year, colleges receive a flurry of inquiries from potential students about all aspects of college and academic life, financial aid and scholarships. To create the best first impression with students and a great user experience, it’s critical for schools to answer all of these questions quickly and accurately.

How the students are treated – and whether they feel well serviced or neglected – can make or break their overall feeling about a university and whether they want to apply or reject the school. A chatbot on a university website can be programmed to talk about what makes the school so unique and why it would be a great place to go, enabling students to click on a topic to learn more.

Once they decide to apply, students typically have a lot of questions. Chatbots can help guide them through the admissions process, which often involves an extensive application with lots of supporting materials. Similarly, they can get help with the many questions they typically have when applying for financial aid.

But even after students apply, are accepted and decide to attend, the process is still not over. Universities lose some of the incoming class to “summer melt,” when accepted students change their minds and don’t end up attending the school. Georgia State University decided to develop a chatbot that could reach out to students who had enrolled via text messages with reminders and key information during the summer before they attend. It also answered questions about the dorm, financial aid, tuition, etc. The school found that the chatbot reduced the summer melt by over 21 percent compared to students in a control group.

Continuing a high-level user experience throughout college

Helping students during the application process isn’t enough. It’s important to maintain a high level of customer service throughout the student’s college career to ensure a positive experience at the school. Studies have shown that when students feel unsupported by the institution it can be a factor in causing them to drop out. Chatbots can be used in in all areas – answering questions about university services such as health services, and athletic services, clubs, student accounts and other aspects of student life.

To answer the growing number of student questions, the Inter-American University of Puerto Rico developed a chatbot that would communicate with students where they predominately spent their time — on the website, and via Facebook Messenger and SMS mobile. The university found that it was able to automate responses for 80 percent of the calls, while reducing by half the support staff needed for this task. Because their time was freed up from having to handle all the calls, the staff could focus on answering more complex questions and proactively help students meet financial aid deadlines, select courses course selection and get any academic help they might need.

Chatbots can also be used for technical support, answering some of the routine questions, so the support team can be more available to focus on the more difficult issues.

Enabling consistency and accuracy

Through years of experience fielding questions from prospective and incoming students, universities can easily develop a list of commonly asked questions, and prepare answers that are clear, accurate and consistent. Chatbots can be programmed to answer these routine questions quickly, while routing more complex questions to human staffers. They also provide a high degree of quality control. Since you program in all of the answers, you can ensure that they are consistent and accurate, something that is harder to do with humans, as well as with websites, that require constant updating.

Every industry is interested in data today and higher education is no exception. Since chatbots answer lots of student queries, they provide a hotbed of documented data and insights into student behavior, student concerns, and areas where the university is strong and where it may fall short. By providing a sort of transactional record of each transaction, chatbots enable universities to use patterns of data to improve their operations. For example, are many student asking questions that should be standard information on the website? Perhaps the website needs updating, with clearer information. Are many students voicing concerns over library closing times? Maybe the hours of operation need changing.

Chatbots provide an easy, cost-effective solution to help universities overcome some of their most pressing challenges today, like declining enrollment and summer melt. It’s quickly becoming apparent that with the help of chatbots, the schools that can best address student needs quickly, effectively and accurately, will win over their hearts and minds – and be well positioned to succeed in today’s competitive marketplace.

Data Science Today is About Much More Than Data or Science

I recently wrote an article for Forbes that looked at what it takes to be a good data scientist these days. With the huge pace of innovation taking place in AI it’s not only difficult to keep up with a changing industry, but it requires skills today that were not even discussed even five years ago.

The Forbes article looked at courses that are required today, such as fundamentals of Hadoop or Apache Spark, as well as machine learning, data visualization and the standard mathematics, statistics, computer science and engineering classes, but it discusses the critical role of the “softer” skills. As the article states, “Data science is all about human interactions, teaching software to think like humans. In fact, Stanford University offers its computer science students classes in persuasion – how to persuade consumers or customers to buy certain things, buy in to your messages and then build those techniques into the software.” As AI tools take on the role of humans and think like humans in many cases, humans need to learn how to help them take on that role.

What’s key to being an effective data scientist is knowing how software interacts with people –
skills that haven’t always come from math or science-based classes. These skills have been taught in more liberal arts focused studies, such as English, sociology, psychology or even history. It’s wise for today’s rising data scientists to become well-rounded, stepping away from the computer lab and studying the humanities as well.

The article also shares how it’s no longer enough to be a really good programmer – in fact within five years most programming will be done by machines. But what’s key is understanding the business – its challenges, goals and customers, as well as having the ability to communicate and solve problems across departments.

There’s amazing new opportunities for students seeking careers in data science, and by focusing on expanding human skills, along with the scientific ones, they’ll be well-prepared to help software become human-like, instead of the other way around.

How is AI Humanely Attacking the Deadliest Animal on Earth? NVIDIA Article Sheds Light

Our Innovation Director, Leslie, recently was interviewed for an article that appeared in the NVIDIA blog, that discusses Wovenware’s work developing an AI solution to help with the identification and classification of mosquitoes carrying deadly diseases. The deep learning solution is being developed for the Puerto Rico Vector Control Unit (PRVCU), to aid in its project to control the spread of diseases across the island.

As the article states, “Ask folks which animal kills the most people each year and they’ll probably say crocodiles, sharks or maybe lions. But the correct answer is a lot less obvious — the mosquito.” Thanks to the PRVCU’s hard work, maybe one day mosquitoes will lose this title and go back to being simply annoying.

We’re very proud of the fact that our deep learning solution is automating the identification and classification of Aedes Aegyptis, which is infecting people with diseases such as Zika, Dengue and Chikungunya across Puerto Rico and nationwide. The other purpose of the project is to develop safe and more effective insecticides.

Our solution is built upon convolutional neural networks (CNNs), which enable it to perform gender and species classification. The two classifier CNNs can then be consolidated into one for training. Leslie said the team is also experimenting with TensorFlow to further speed the inference process.

The article explains that while there are other initiatives underway to combat mosquito-borne diseases, Wovenware is more concerned with technology solutions that would provide the least disruption to the food chain. As Leslie says, “Every species has a reason to live,” and I don’t see why we should need to extinguish a species.”

We’re very grateful to NVIDIA for shedding light on how deep learning AI can help address some of the most pressing environmental and health-related concerns. Stay tuned for more updates as our AI solution continues to do its part in helping to eradicate mosquito-borne diseases.

Getting Started with Mobile Development

The idea of having a small gadget in the palm of your hand, helping you with your daily tasks, organizing your thoughts, giving you a digital calendar to manage your busy schedule, allowing you to send and receive phone calls through different mediums, and use a high resolution camera, along with fun games and other entertainment on the go, has intrigued me all my life.

The truth is that mobile apps have become a big part of our life today. In the US, people use on average 9 apps a day, about 30 apps monthly, and spend an average of 3 hours and 35 minutes daily on mobile devices. These facts are what got me interested in the world of mobile apps.

My name is Pablo, I’m currently 27 years old, and I’ve been programming for more than five years now, but I’ve been interested in software development since the first time I used a computer. I’ve always wondered how things work behind the scenes. Event at a young age, I would find myself searching the web on how to create my own app.  Yet, seeing all of those unfamiliar terms, names, programming languages, etc., I got discouraged because I didn’t know where to get started. That’s why I wanted to write this article. To show how I got started, and help anyone else considering a career in mobile app development.

It is worth noting that before learning how to develop mobile apps, you will need a good grasp of software development skills and knowledge of key programming language, such as JAVA, C or C++. It might seem a little intimidating at first, but there are a diverse amount of resources to get you started. You can find multiple online courses, for example, on Udemy, Coursera, Pluralsight involving software development. Obviously if you are in college taking anything remotely close to software development (management information systems, for example) as a degree, you will most certainly gain these skills and knowledge. I learned these basics in college, as well as on my own using online courses from different sources.

One of the first decisions I had to make was if I wanted to develop on iOS or Android. Even though these are not the only platforms that we can develop apps on, they are definitely the most globally popular and their established communities help us find the resources we need to answer any questions we may have on the way.

It’s important to make this decision since the resources needed, such as Software Development Kits (SDKs) or Integrated Development Environments (IDEs), vary for each targeted operating system you select. There’s also the difference of working environments between the two operating systems. When developing for iOS, the environment is a bit limited. In contrast, when developing for Android, the environment is more flexible. Keep in mind that the minimum recommended specs for the computer you use is not intended for mobile development specifically, but overall software development. While you don’t need the strongest computer, a good computer helps with compilations and build times.

Minimum requirements for developing iOS apps:

  •  An apple computer with MacOS is needed
    • Intel i5 or i7 equivalent CPU, ~2.5 Ghz
    • 4GB of RAM, although I would recommend 8GB
    • 128 GB disk storage
  • xCode IDE (This is free to download from the apple App Store)

Minimum requirements for developing Android apps:

  • You can use MacOS or Windows with more or less the same specs
    • Intel i5 or i7 equivalent CPU, ~2.5 Ghz
    • 4GB of RAM, although I would recommend 8GB
    • 128 GB disk storage
  • Android Studio IDE (This is free to download from the web)

Once you have decided which platform and environment you will work on, you need to consider the programming languages available for developing in these specific platforms natively. iOS development has two different programming languages, Swift and Objective-C. Objective-C is a much more mature programming language, having been around for at least 30 years. It was the first one available for users to develop iOS mobile applications. In 2014 Swift was introduced. Swift aims to make things easier, from syntax to memory management. Swift has now been evolving rather rapidly, especially when it was open-sourced by Apple in 2015. I started iOS development using Swift and if you are just starting, I highly recommend it. While Objective-C is still available and being used, it is reasonable to say that Apple is focusing on the future of Swift. Here are some resources to get started on iOS development:

Android development can also be written on two different languages, Kotlin and JAVA. Although most of my experience on mobile development lies within iOS development, it is my understanding that learning Android development should start off using JAVA. This is because learning JAVA along the way helps you to advance career wise and become part of a large community, which could lead to better job opportunities. At the same time, if you are already pretty experienced with JAVA, learning Kotlin will help you prepare for what’s to come in the future and also improves productivity considering that there are some tasks that are easier in Kotlin than on JAVA. Here are some resources to get started on Android development:

Regardless of the resources you leverage, the biggest step to take is getting started. Software development skills are not only in high demand but are increasingly becoming an essential skill for everyone. Once you have learned the basics in software and mobile development, you can ignite your creativity and develop awesome apps whether just for fun, or to help solve business needs. Happy coding!