Daily Tech Digest - March 12, 2018

Monetizing Services in Digital Time with AI and Machine Learning

Machine learning can help determine the right level of personalization for offers and deliver them at the perfect digital moment, increasing take-up rates for services. AI can identify problems faster with automated root cause analysis and anomaly detection, then trigger corrective actions to ensure retention and protect revenues. According to eMarketer, in 2018, around 1.87 billion individuals worldwide will use a mobile phone to watch digital video, an 11.9 percent increase compared to 2017. With real-time offerings like this on the line, every second counts when solving problems. Machine learning even brings opportunities to monetize operator data towards third parties, with offerings for verticals like retail and transportation. Combined with demographics and social media, operators can sell reports to show anonymized movements of crowds for tracking store performance against competitors, choosing the next location for a shop, seeing what are the most traveled routes for urban planning, and deciding how to lay out the next metro line.


Ransomware for robots is the next big security nightmare

robot-ransomware.jpg
Taking what was learned in previous studies into the security vulnerabilities of robots, researchers were able to inject and run code in Pepper and NAO robots and take complete control of the systems, giving them the option to shut the robot down or modify its actions. The researchers said it was possible for an attacker with access to the Wi-Fi network the robot is running on to inject malicious code into the machine. "The attack can come from a computer or other device that is connected to internet, so a computer gets hacked, and from there, the robot can be hacked since it's in the same network as the hacked computer," said Cerrudo. Unlike computers, robots don't yet store vast amounts of valuable information that the user might be willing to pay a ransom to retrieve. But, as companies often don't have backups to restore systems from, if a robot becomes infected with ransomware, it's almost impossible for the user to restore it to normal by themselves.


How Postgresql Just Might Replace Your Oracle Database

PostgreSQL replacing Oracle database? Salesforce might make it happen
If, in fact, Salesforce is developing a homegrown replacement for Oracle’s database, it might well be building it on PostgreSQL, the database Salesforce has actively flirted with since 2012. In 2013, Salesforce hired Tom Lane, a prominent PostgreSQL developer. In that same year, it hired several more, and even today PostgreSQL experience is called out for in dozens of jobs advertised on the company’s career page. Just as Facebook, Google, and other web giants have shaped MySQL to meet their aggressive demands for scale, so too might Salesforce be able to mold PostgreSQL to wean it from its dependence on Oracle. ... Oracle would claim that it isn’t worried, but the DB-Engines database popularity ranking, which measures database popularity across a range of factors, should give it pause. For years, PostgreSQL has been on the rise, even as Oracle and MySQL  have faded. PostgreSQL is now a strong fourth place, with MongoDB right behind it.


The 10 most common cybersecurity scams uncovered

In the beginning there was the internet, and shortly after that came the internet scammers. Online scams include everything from the now-legendary Nigerian prince meme to the less-well-known but infinitely more devious fake shopping websites. If you were curious about the origins of these deceitful hoaxes, we’ve got you covered. On the other hand, should you be worried about the repercussions of falling victim to one of these fraudulent schemes, we’ll also touch on that.  Online scams are typically malware disguised as rewards or charitable gestures. After all, what is the Nigerian prince scam other than an attempt to get you to care about getting someone else out of a rut, and providing you with a huge payout for doing so? All the examples below are designed to prick our consciences, or play on our greed or vanity in one way or another. Some of these cyber-scams are actually pretty ingenious, but ultimately malicious – others are just plain malicious.


Could Singapore hold the secret to preparing workers for an uncertain future?


Singapore offers a simple yet elegant solution: “second-skilling.” Tay realized that in today’s economy, second-skilling — developing your skills in a sector other than the one you work in — is necessary for career resiliency; it gives you options and flexibility. That second skill can either complement the skills you’re already using in your current job, or offer a completely alternative path. But who pays for second-skill training? The answer in Singapore is surprising. Thanks in part to Tay’s lobbying, every Singaporean 25 and older gets S$500 (about US$350) for skills training of their choice from the government through the SkillsFuture program. The money’s in a virtual credit account, and the government plans to provide periodic top-ups. It can pay for training in anything a person might want to learn, not just what their company needs them to know. “Many programs are already funded 80 to 90 percent,” says Tays. “So the five hundred dollars can be used to pay for the unfunded portions, which, previously, we had to fork out from our own pockets.”


5 emerging tech trends at SXSW that will shape 2018

Sadly, generalized intelligence (think Hal 9000) is still a dream in engineer’s mind. However, that doesn’t mean that AI isn’t still a hot topic for 2018. With computers getting faster and GPU’s being re-purposed, we’re seeing an explosion of innovation, from machine-learning models to validate brand creative to those that create unique art and music. Companies like IBM and Getty Images are asking how they can apply large scale AI to the creative process, and what that means for their business. Meanwhile, L’OrĂ©al is applying machine-learning to improve product recommendations through its smart hairbrush. Despite all that, some of the more compelling topics this year are around the ethics and morality of AI. As AI is used in more serious applications (like self-driving cars and medicine), we rely on machines to make life-or-death decisions. Who is responsible for these decisions, and what rules do they follow? Still a Wild West to be figured out.


Third-party security vetting: Do it before you sign a contract

partnership collaboration puzzle pieces unity
Archer contends relationships are key. “Security has to be able to say, ‘We're not going to do business with that vendor,’” he says. To enforce a policy like that, the c-suite must take security seriously. If there’s not a CSO to represent you, talk to the CEO yourself. “If you can't get through the front door, maybe you get through the back door,” he recommends. Either way, he adds, “Establish those relationships.” Then grow relationships with the actual prospective vendors. At Fannie Mae, this starts with a security best practices questionnaire included in all RFIs. Archer’s team divided vendors into two groups — critical and regular — by the type of data they’ll access. For prospective critical vendors, there are around 250 questions. Regular vendors get shorter, industry-specific versions of the questionnaire. Most questions for both groups are primarily yes or no: “Are you SOC 1 and SOC 2 compliant?”, for example. The RFI is also an opportunity for prospective vendors to get to know you. In addition to adding questions, Fannie Mae outlines security expectations.


How Retail Shifted from Business Intelligence to Data Science

data science business intelligence retail
Being competitive in retail often means raising or lowering prices depending on what rival companies are also offering. Walmart is one retailer that spends tons of money on monitoring not just its own transactions, but also the price changes of its competitors while taking into consideration its own stock levels. With the aid of data science, Walmart is able to implement real-time changes to its pricing and never lose its edge over other retailers. In the past, to price products this fiercely took a lot more time and couldn’t be monitored so conveniently or, most importantly, predicted. This also has a second advantage, which is the possibility of moving away from the traditional end of sales technique of significantly lowering prices once demand has almost vanished at the end of a season, to instead drop prices more gradually, which has been shown to be more effective. Business intelligence is still an incredibly useful thing to have, with 62% of retailers reporting that using data is creating a competitive advantage, according to IBM.


Aryaka adds new security layers to global private network


Each security layer, which ranges from distributed denial-of-service protection to network edge and cloud security options, offers specific features, according to Gary Sevounts, chief marketing officer at Aryaka, based in San Mateo, Calif. Additional layers include a virtual firewall for Amazon Web Services and Microsoft Azure instances and built-in compartmentalization and contained environment controls. Finally, an early warning portal identifies behavior anomalies and potential risks. By offering multiple security layers from different security vendors, Sevounts said Aryaka's customers can receive extra protection in case there is a problem with one of the layers. "If you have all your layers from a single provider and something happens to the core technology, then all the layers are ineffective, and the company is susceptible to attacks," he said.


A new mindset drives a new way forward

collaboration brainstorming mobile [RawPixel.com - CC0 via Unsplash]
Why should IT have to struggle deploying cobbled-together kits, and trying to overcome difficult lighting or acoustic challenges? And why should participants have to spend 10 minutes trying to launch a meeting? Or suffer through inaudible sound, shouting into phones, camera angles that are too tight or too wide for the room, etc.? So yes, more scenarios are being supported, but the result is greater complexity to achieve a mediocre (or worse) experience — and this frustrates people, hampers productivity, and slows adoption. To make matters worse, those solutions can come at an unnecessarily high cost to buy and manage. In order to make real strides in collaboration today and into the future, you have to simultaneously address all three competing pressures. Solutions have to be able to support more scenarios — including scenarios we haven’t yet imagined — in simpler ways, while delivering a better experience.



Quote for the day:


"The past has no power over the present moment." -- Eckhart Tolle