AI and machine learning require expert data scientists, engineers, and researchers, and there's a worldwide short supply right now. The ability of autoML to automate some of the repetitive tasks of ML compensates for the lack of AI/ML experts while boosting the productivity of their data scientists. By automating repetitive ML tasks -- such as choosing data sources, data prep, and feature selection -- marketing and business analysts spend more time on essential tasks. Data scientists build more models in less time, improve model quality and accuracy, and fine-tune more new algorithms. More than 40 percent of data science tasks will be automated by 2020, according to Gartner. This automation will result in the increased productivity of professional data scientists and broader use of data and analytics by citizen data scientists. AutoML tools for this user group usually offer a simple point-and-click interface for loading data and building ML models. Most autoML tools focus on model building rather than automating an entire, specific business function such as customer analytics or marketing analytics.
Humans are, well, only human. Mistakes happen, but a mistake can have a huge impact on an organisation’s health and future success. Introducing technology to manage a range of processes can help to reduce and mitigate HR related risk by minimising all manner of issues from poor HR consistency and visibility, to data loss. Manually updating changes in spreadsheets can be a cumbersome and ineffective process, especially when the data is being entered into multiple documents. Research from Salesforce shows that 88% of all spreadsheets have significant errors in them. Applying intelligent automation will not only reduce the risk of human mistakes but also help to flag errors and data problems before they create a negative impact on the business. The huge issue of risk and compliance aside, automation reduces the HR admin mountain and allows a focus on people strategies which are so critical when competing for talent and reducing churn.
While many of you may see edge as exclusive to IoT, its value is much wider and will prove as critical to driving up customer experience as content delivery networks (CDN) were in the early days of the web . . .which explains why you are now seeing edge compute and AI services from all the major cloud vendors and on the road maps of the leading telecom companies. Twenty-seven percent of global telecom decision makers, who responded this year to the Forrester Analytics Global Business Technographics® Mobility Survey, 2018, said that their firms are either implementing or expanding edge computing in 2019. Many of these vendors will require new wireless tools and updated skill sets to achieve this digital transformation. This aligns to Verizon's recent employee buyout offer, as a result of which over 10,400 of its staff will be gone next year, driving nearly $10 billion in savings that it can apply to its edge-compute-empowered 5G network. And speaking of CDNs, nearly every one of these vendors is adding edge compute to their core market values.
The story highlights the shortcomings of purportedly “state of the art” AI automation that are rarely discussed. One is that they’re installed to solve a management problem rather than a customer need, as was the case here - the hotel is in an area with an acute labour shortage. Secondly, they’re just plain annoying. As hotel manager Hideo Sadawa explained: “When you actually use robots you realize there are places where they aren’t needed - or just annoy people”. While robotics has advanced steadily in industry, the picture is different in consumer electronics. Trade group the International Federation of Robotics noted that sales of industrial robots had doubled in five years. But it’s largely cyclical, IFR president Junji Tsuda admitted. Adoption doubled even more dramatically between 2009 and 2010, which had nothing to do with AI and a lot to do with the falling cost of sensors and microelectronics. In industries where automation is highly advanced, such as car production, it may not move the dial much: wage rates largely govern the substitution phenomenon
Hackers continue to prefer phishing schemes to almost any other infiltration or social engineering tactic. In part, their effectiveness ties into their mundanity; phishing attacks look like legitimate emails, and employees without proper training will reliably open their emails. Phishing attacks, therefore, provide a low effort, high impact cyber threat. Furthermore, if it can hit the SEC, it can hit your enterprise as well. To prevent a phishing attack from inflicting damage on your databases, make sure your employees can recognize a phishing attack if they receive one; there are tell-tale signs for almost all of them. Incentivize recognizing phishing attacks before they occur, either through a small rewards program or by making cybersecurity a part of your employees’ everyday job duties and performance reviews. Additionally, ensure your cybersecurity platform includes a SIEM solution with strong threat detection capabilities. Your enterprise can also benefit from an email security solution to prevent phishing attacks from reaching your inboxes.
With the PNR and customer name at our disposal, we were able to log into ELAL’s customer portal and make changes, claim frequent flyer miles to a personal account, assign seats and meals, and update the customer’s email and phone number, which could then be used to cancel/change flight reservation via customer service. Though the security breach requires knowledge of the PNR code, ELAL sends these codes via unencrypted email, and many people even share them on Facebook or Instagram. But that’s just the tip of the iceberg. After running a small and non-threatening script to check for any brute-force protections, none of which were found, we were able to find PNRs of random customers, which included all of their personal information. We contacted ELAL immediately to point out the threat and prompt them to close the breach before it was discovered by anyone with malicious intentions. We suggested stemming the vulnerability by introducing captchas, passwords, and a bot protection mechanism, in order to avoid using a brute-force approach.
New concepts and terminology have been introduced in the COBIT Core Model, which includes 40 governance and management objectives for establishing a governance program. The performance management system now allows more flexibility when using maturity and capability measurements. Overall, the framework is designed to give businesses more flexibility when customizing an IT governance strategy. Like other IT management frameworks, COBIT helps align business goals with IT goals by establishing links between the two and creating a process that can help bridge a gap between IT — or IT silos — and outside departments. One major difference between COBIT and other frameworks is that it focuses specifically on security, risk management and information governance. This is emphasized in COBIT 2019, with better definitions of what COBIT is and what it isn’t.
The report on the security analysis of radio remote controllers for industrial applications highlights notes the use of obscure, proprietary protocols instead of standard ones makes controllers vulnerable to command spoofing, so an attacker can selectively alter their behaviour by crafting arbitrary commands, with consequences ranging from theft and extortion to sabotage and injury. “The legacy and widespread RF technology used to control industrial machines is affected by serious security issues that impact several market verticals, applications, products and brands,” the report said. The researchers warned that currently and widely used legacy RF technology for industrial applications can be abused for sabotage of equipment, theft of goods by manipulating equipment and extortion by demanding payment to hold off or cease equipment interference.
Quote for the day:
"Your talent and giftedness as a leader have the potential to take you farther than your character can sustain you. That ought to scare you." -- Andy Stanley