Half of the worlds VPNs would not be secure – InTallaght

Posted on April 26, 2020 by admin

The VPN they are essential tools to avoid internet censorship, as well as to browse safely encrypting our entire connection and hiding it from the eyes of authorities and operators. However, many of the VPNs would not be secure, a new report reveals.

The report states that in the data they have collected, the 46.6% of all VPNs of the world are based in countries that participate in the pact of 14 eyes, where 14 countries of the world participate sharing intelligence, among which is Spain. This group includes, in turn, the Five Eyes and Nine Eyes, formed by:

Members of these agreements may share intelligence data, and you can even skip the laws that prohibit citizen surveillance. Various VPNs warn of the possibility of this type of situation, and therefore warn that using one of a specific country can lead to security problems.

Thus, a VPN could be forced to share user information with the government, which in turn would allow any member of the 14 countries to obtain such information for being part of the intelligence pact. For example, the United States might know the browsing history of a user who has committed a crime.

The eyes pact began with five countries in the 1940s and aimed to share information of military origin in the face of the Cold War. However, little by little, it was expanding to intelligence, and now it also affects the information that is available online.

The Fourteen Eyes group is also better known as SIGINT Seniors Europe, or SSEUR, according to documents leaked by Edward Snowden. The relationship between the countries that make it up is not as close and intimate as that of the Five Eyes, but even so, they can share sensitive information belonging to users.

Thus, 46.6% of VPNs are present in one of those 14 countries, and the figure increases to 48.4% if we consider that there are countries that also collaborate at the intelligence level such as Israel, Singapore, Japan and Korea from the south. The risk is even higher by users using free VPNs, most of whom do not guarantee 100% anonymity; especially if a judge asks them to identify a user.

Among the data collected by VPNs are the websites visited, connection times, bandwidth, server location, and even the original IP address with which they can contact an operator to help them identify the real user with the name. And surnames.

Therefore, the key is to use a VPN that does not keep records, and that is located in a country that does not fall within the Fourteen Eyes alliance. For example, Express VPN is based in the British Virgin Islands, while Nord VPN It is found in Panama, so they would be the best VPNs that we can use at the privacy level.

Go here to read the rest:
Half of the worlds VPNs would not be secure - InTallaght

Zoom Upgrades Encryption Keys to What It Promised All Along – WIRED

Posted on April 26, 2020 by admin

It was another week of social distancing or quarantine for most of the world, but Google published findings that it has seen 12 government-backed hacking groups undeterred by the pandemic and, in fact, trying to take advantage of those conditions for intelligence-gathering. Another report found that China, for one, has been busy during the pandemic hacking Uighurs iPhones in a recent months-long campaign.

We broke down how Apple and Google are using aggregate smartphone location data to visualize social distancing trends. And in an exclusive interview with WIRED, Federal Bureau of Investigation director Christopher Wray warned that domestic terrorism is a growing threat in the United States.

On top of all the other digital threats, researchers emphasized this week that so-called "zero-click" hacks that don't require any interaction from users to initiate may be more prevalent and varied than most people realize. Such attacks are difficult to detect with current tools.

And there's more. Every Saturday we round up the security and privacy stories that we didnt break or report on in depth but think you should know about. Click on the headlines to read them, and stay safe out there.

On Wednesday, the video conferencing service Zoom announced a number of small but needed security improvements. As Zoom usage has increased during the pandemic, so has scrutiny on the service's security and privacy offerings. This week's announcement of incremental improvements is part of a 90-day plan the company announced to overhaul its practices. One change is that Zoom will now offer AES 256 encryption on all meetings, meaning data will be encrypted with a 256-bit key. Zoom previously used AES 128, a reasonable option, but a controversial one in Zoom's case, because the company claimed in documentation and marketing materials that it used AES 256 all along.

Facebook data from more then 267 million profiles is being sold on criminal dark web forums for 500, or about $618. The information doesn't include passwords, but does include details like users' full names, phone numbers, and Facebook IDs. Though such information can't be used to break into the accounts directly, it can fuel digital scams like phishing. Most of the trove seems to be the same as data found by researcher Bob Diachenko in an exposed cloud repository last month. Even after that bucket was taken down, though, a copy of the information plus an addition 42 million records popped up in a different repository.

A growing number of Nintendo users over the past few weeks had watched fraudsters take control of their accounts, and in many cases use saved credit cards or linked PayPal accounts to buy Nintendo games or currency for the popular game Fortnite. At the beginning of April, Nintendo encouraged users to turn on two-factor authentication to protect their accounts, but it had been unclear how hackers were breaking in. On Friday, the company confirmed that hackers had gained unauthorized access to accounts and announced it was discontinuing users' ability to log into their Nintendo Accounts using Nintendo Network IDs, from older Wii U and 3DS systems. Nintendo also says it will contact affected users about resetting passwords. On its US customer support page, the company writes, "While we continue to investigate, we would like to reassure users that there is currently no evidence pointing toward a breach of Nintendos databases, servers or services."

Review of the iStorage datAshur Pro2, an encrypted thumbdrive for home and work – Neowin

Posted on April 26, 2020 by admin

With all of the data leaks being announced on a daily basis, many people have decided that they would rather host their own data in order to maintain control. While there are many options, such as NAS devices, sometimes you may want access to your files anywhere you are, regardless of an Internet connection, so a portable storage device is something to consider. But once the device leaves the house, what happens if you accidentally lose it or your laptop bag is stolen?

Back in 2018, I took a look at the iStorage diskAshur PRO2, a secure portable hard drive that encrypted all data on the device. Today, I'll take a look at the iStorage datAshur PRO2, a device that provides similar functionality to the diskAshur PRO2, only in a small USB thumbdrive form factor that you can easily put in your pocket and carry with you.

The datAshur PRO2 comes in different capacities, from 4GB (priced at 49/$59) all the way up to 512GB (priced at 369/$468). Our review unit, provided by iStorage, was the 256GB model, priced at 279/$358.

3.7V Li-Polymer Rechargeable Battery

USB 3.2 Gen 2 SuperSpeed USB

The device's housing is made of a rugged anodized aluminum, and the case that goes over it has a rubber gasket to make the datAshur PRO2 waterproof, giving it an IP58 certification.

The USB drive contains a rechargeable battery. This lets you type in the PIN and unlock the device before connecting the datAshur PRO2 to a USB port on your computer. Since all of the encryption and decryption is done on the thumb drive itself, the datAshur PRO2 can be used on Windows, Mac, and Linux without the need to install any drivers. If the battery dies from lack of use, you can still plug it into a computer and unlock the drive while it's charging.

If you're paying for an encrypted drive, part of the cost is related to obtaining security certifications, and the datAshur PRO2 has a lot of them. Or rather, is in the process of obtaining a lot of them. Unfortunately it's hard to tell which ones have been completed and which ones are currently in process. Let's explore those certifications in the next section.

The buttons on the device are the main interface, and they have a polymer coating to prevent the keys from smudging, which would tip off an attacker as to what keys are frequently pressed to unlock it.

If you're looking for cheap and portable storage, but don't care about security, then there are far better solutions in the market for you that come in at much cheaper price points. If you require your data to be secure, then the datAshur PRO2 shines.

The website lists the following certiifcations/validations for the datAshur PRO2: FIPS 140-2 Level 3, NCSC CPA, NLNCSA BSPA & NATO Restricted Level. However it's hard to tell which of these the product has completed, and which are still pending review by the official agencies. For example, the device does not yet have FIPS 140-2 Level 3 validation, although the design itself is compliant. If this distinction is important to you for purchasing, you'll want to wait a little longer for NIST to complete its review. The company has other products that have been approved, so I suspect it's only a matter of time before the datAshur PRO2 is validated as well.

One of the main physical features that will allow iStorage to obtain FIPS 140-2 Level 3 validation on the datAshur PRO2 is the fact that the components within the device are covered in an epoxy resin. This makes physical attacks against the hardware nearly impossible without damaging the components, and is also used as evidence that the device has been tampered with.

The datAshur PRO2 is a great solution for enterprise environments due to the ability to whitelist the device. This allows a company to lockdown what external storage devices are allowed to connect to the corporate network and prevent someone from, for example, connecting a rogue USB thumbdrive that may have malware on it, while allowing approved devices. Unfortunately, my test lab does not have the ability to test this functionality.

The device has a concept of both an admin and a regular user account. The admin account allows you to setup features on the datAshur PRO2, such as password complexity, create a one-time recovery PIN, set the device as bootable, or make the device read-only to users. The user account only allows someone to read/write data stored on the device, and of course, only the admin can create users.

Another interesting feature of the datAshur PRO2 is the concept of a "self-destruct" PIN. If configured, you can simply type in the code to automatically delete the encryption keys and the data from the device. While this feels very spy-like, it's better to have the feature and not need it than need it and not have it.

While the datAshur PRO2 has a lot of features available, you will probably want to keep the manual nearby because memorizing all of the keypresses required to set things up or check the status is pretty much impossible. However, the PDF, which is included on the drive itself when you first get buy it, is very informative and has clear instructions on what to do and how the device will respond, as shown in the example above. In addition, unlocking the device is what you'll be doing the most, and that's a simple process to perform.

Since the datAshur PRO2 supports a USB 3.2 Gen 2 SuperSpeed connection, I expected the performance to be great and I was not disappointed.

While I normally test NAS systems using a RAM-disk in order to reduce latency, I went with a more "real-world" test here of simply copying files my Samsung 970 EVO Plus NVMe drive to the datAshur PRO2. Since the drive can read and write at over 3,300 MB/sec, we'll definitely only be limited by the USB bus and the USB drive itself.

According to the specs on the iStorage website, you can expect a maximum of 168 MB/sec when reading files from the device, and 116 MB/sec when writing files to to the device.

After doing my normal battery of tests - namely, copying a multi-gigabyte file to/from the device and copying hundreds of multi-megabyte files to/from the device, I was roughly hitting those numbers. When connected to a USB 3.0 or better port, a simple drag-and-drop copy would always consistently show 116 MB/sec when writing files to the datAshur PRO2 and 130 MB/sec when reading from the datAshur PRO2. When connected to an older USB 2.0 port, performance was consistently throttled at 40 MB/sec for all of the tests.

Running a benchmark in CrystalDiskMark provided even higher throughput values, with a consistent 140 MB/s sequential read and 128 MB/s sequential write speeds. As expected, random read/write speed was considerably slower, clocking in at only 17 MB/s read and a maximum of 12 MB/s write.

While it would've been nice to see the device take full advantage of the USB 3.2 performance, these speeds are still reasonable and iStorage delivers what it promises.

Unlike a regular flash drive, if you plug the datAshur PRO2 into a USB drive, nothing happens. In order to access the drive, you have to first unlock it. This process consists of holding the shift key down on the drive to wake it up, then pressing the unlock key, followed by your PIN code. The red light under the locked padlock lights up when the datAshur PRO2 is awake, and the unlocked padlock and letter A (administrator) start flashing after you press the unlock key. Once unlocked, you have 30 seconds to plug the device into a computer before it automatically locks itself again. Alternatively, you can plug the device in first and then unlock it.

The keys themselves are only slightly raised, but have a nice tactile response when pressed, making it easy to know you've successfully engaged it. In all of my testing, I never once accidentally pressed the wrong button.

When in the unlocked mode and connected to a computer, the datAshur PRO2 stays unlocked until it's removed from the USB port or the computer is rebooted. During this time, it acts just like a regular flash drive you'd connect to your PC.

The one minor quibble is that when you pull the flash drive from its protective sleeve, you have to make sure you don't misplace it. The datAshur PRO2 is only waterproof when in the sleeve, due to the rubber gasket, and the sleeve also protects against accidental button presses.

There's an old adage in technology: Good, Fast, Cheap, pick two. It's clear that iStorage went in with the strategy of creating a good and fast product, but as I noted earlier, that means the device is not cheap, especially at the larger capacities.

If you're just looking for a USB flash drive to store some generic data, then you probably won't be interested in the datAshur PRO2. It's a little bigger than a normal drive and a lot more expensive, especially as you look at the larger capacity drives. However if you're looking for a well designed, well performing, and very secure thumb drive to store sensitive data, you can't go wrong with the datAshur PRO2. In addition, if you're only looking to keep some sensitive documents that don't take up a lot of space, the four gigabyte version is only $59, which is a great price considering the level of security the device maintains.

More:
Review of the iStorage datAshur Pro2, an encrypted thumbdrive for home and work - Neowin

Encryption Software Market Segmentation, Application, Technology, Analysis Research Report and Forecast to 2026 – Cole of Duty

Posted on April 26, 2020 by admin

Microsoft Corporation

Global Encryption Software Market Segmentation

This market was divided into types, applications and regions. The growth of each segment provides an accurate calculation and forecast of sales by type and application in terms of volume and value for the period between 2020 and 2026. This analysis can help you develop your business by targeting niche markets. Market share data are available at global and regional levels. The regions covered by the report are North America, Europe, the Asia-Pacific region, the Middle East, and Africa and Latin America. Research analysts understand the competitive forces and provide competitive analysis for each competitor separately.

To get Incredible Discounts on this Premium Report, Click Here @ https://www.verifiedmarketresearch.com/ask-for-discount/?rid=1826&utm_source=COD&utm_medium=005

Encryption Software Market Region Coverage (Regional Production, Demand & Forecast by Countries etc.):

North America (U.S., Canada, Mexico)

Europe (Germany, U.K., France, Italy, Russia, Spain etc.)

Asia-Pacific (China, India, Japan, Southeast Asia etc.)

South America (Brazil, Argentina etc.)

Middle East & Africa (Saudi Arabia, South Africa etc.)

Some Notable Report Offerings:

-> We will give you an assessment of the extent to which the market acquire commercial characteristics along with examples or instances of information that helps your assessment.

-> We will also support to identify standard/customary terms and conditions such as discounts, warranties, inspection, buyer financing, and acceptance for the Encryption Software industry.

-> We will further help you in finding any price ranges, pricing issues, and determination of price fluctuation of products in Encryption Software industry.

-> Furthermore, we will help you to identify any crucial trends to predict Encryption Software market growth rate up to 2026.

-> Lastly, the analyzed report will predict the general tendency for supply and demand in the Encryption Software market.

Have Any Query? Ask Our Expert @ https://www.verifiedmarketresearch.com/product/global-encryption-software-market-size-and-forecast-to-2025/?utm_source=COD&utm_medium=005

Table of Contents:

Study Coverage: It includes study objectives, years considered for the research study, growth rate and Encryption Software market size of type and application segments, key manufacturers covered, product scope, and highlights of segmental analysis.

Executive Summary: In this section, the report focuses on analysis of macroscopic indicators, market issues, drivers, and trends, competitive landscape, CAGR of the global Encryption Software market, and global production. Under the global production chapter, the authors of the report have included market pricing and trends, global capacity, global production, and global revenue forecasts.

Encryption Software Market Size by Manufacturer: Here, the report concentrates on revenue and production shares of manufacturers for all the years of the forecast period. It also focuses on price by manufacturer and expansion plans and mergers and acquisitions of companies.

Production by Region: It shows how the revenue and production in the global market are distributed among different regions. Each regional market is extensively studied here on the basis of import and export, key players, revenue, and production.

About us:

Verified market research partners with the customer and offer an insight into strategic and growth analyzes, Data necessary to achieve corporate goals and objectives. Our core values are trust, integrity and authenticity for our customers.

Analysts with a high level of expertise in data collection and governance use industrial techniques to collect and analyze data in all phases. Our analysts are trained to combine modern data collection techniques, superior research methodology, expertise and years of collective experience to produce informative and accurate research reports.

Mr. Edwyne FernandesCall: +1 (650) 781 4080Email: [emailprotected]research.com

Tags: Encryption Software Market Size, Encryption Software Market Trends, Encryption Software Market Growth, Encryption Software Market Forecast, Encryption Software Market Analysis

Here is the original post:
Encryption Software Market Segmentation, Application, Technology, Analysis Research Report and Forecast to 2026 - Cole of Duty

Global Cloud Encryption Technology Market Expected to Witness the Highest Growth with Gemalto, Sophos, Symantec, SkyHigh Networks and Forecast 2026 -…

Posted on April 26, 2020 by admin

Global Cloud Encryption Technology Market 2019, this report is prepared by in-depth analysis of historical data. The report forecasts the market size by the end of 2026 at an impressive CAGR provided by Reports and Reports. The report offers detailed outline of Cloud Encryption Technology Market and vital market trends. The prime agenda of this report is to provide a detailed analysis of the global, regional and country-level market size, market growth, market status, forecast, sales analysis, value chain optimization, trade regulations, recent developments, opportunities analysis, and importance of the global and national market players, competitive environment, expansion, acquisition, partnerships and technological innovations. The prime market segments considered to prepare this report are key players, regional segments, type and application.

Get Free Sample Copy of Cloud Encryption Technology Market: http://www.milliondollarresearch.com/cloud-encryption-technology-market/928/#RequestSample

The Cloud Encryption Technology market report provides a detailed analysis of global market size, regional and country-level market size, segmentation market growth, market share, competitive Landscape, sales analysis, impact of domestic and global market players, value chain optimization, trade regulations, recent developments, opportunities analysis, strategic market growth analysis, product launches, area marketplace expanding, and technological innovations.

The major players covered in Cloud Encryption Technology Market are: Gemalto, Sophos, Symantec, SkyHigh Networks, Netskope.

Cloud Encryption Technology Industry 2020 Global Market Research report presents an in-depth analysis of the Cloud Encryption Technology market size, growth, share, segments, manufacturers, and technologies, key trends, market drivers, challenges, standardization, deployment models, opportunities, future roadmap and 2026 forecast. The report additionally presents forecasts for Cloud Encryption Technology market revenue, consumption, production, and growth drivers of the market.

Get Special Discount on this report: http://www.milliondollarresearch.com/cloud-encryption-technology-market/928/#AskForDiscount

Regions and Countries Level Analysis:

The important regions, considered to prepare this report are North America (United States, Canada and Mexico), Europe (Germany, France, UK, Russia and Italy), Asia-Pacific (China, Japan, Korea, India and Southeast Asia), South America (Brazil, Argentina, Colombia), Middle East and Africa (Saudi Arabia, UAE, Egypt, Nigeria and South Africa). The region wise data analyses the trend, market size of each regions Cloud Encryption Technology Market.

Table of Content (TOC)

Chapter 1: Product definition, type and application, Global market overview;

Chapter 2: Global Market competition by company;

Chapter 3: Global sales revenue, volume and price by type;

Chapter 4: Global sales revenue, volume and price by application;

Chapter 5: United States export and import;

Chapter 6: Company information, business overview, sales data and product specifications;

Chapter 7: Industry chain and raw materials;

Chapter 8: SWOT and Porters Five Forces;

Chapter 9: Conclusion.

Continue

List of Tables and Figures

Get Detail information of this report: http://www.milliondollarresearch.com/cloud-encryption-technology-market/928/

Conclusively, this report is a one-stop reference point for the industrial stakeholders to get Cloud Encryption Technology Market forecast of till 2026. This report helps to know the estimated market size, market status, future development, growth opportunity, challenges, growth drivers of by analyzing the historical overall data of the considered market segments.

See the original post here:
Global Cloud Encryption Technology Market Expected to Witness the Highest Growth with Gemalto, Sophos, Symantec, SkyHigh Networks and Forecast 2026 -...

Database Encryption Market Segmentation, Application, Technology, Analysis Research Report and Forecast to 2026 – Cole of Duty

Posted on April 26, 2020 by admin

Gemalto

Global Database Encryption Market Segmentation

To get Incredible Discounts on this Premium Report, Click Here @ https://www.verifiedmarketresearch.com/ask-for-discount/?rid=2955&utm_source=COD&utm_medium=005

Database Encryption Market Region Coverage (Regional Production, Demand & Forecast by Countries etc.):

North America (U.S., Canada, Mexico)

Europe (Germany, U.K., France, Italy, Russia, Spain etc.)

Asia-Pacific (China, India, Japan, Southeast Asia etc.)

South America (Brazil, Argentina etc.)

Middle East & Africa (Saudi Arabia, South Africa etc.)

Some Notable Report Offerings:

-> We will give you an assessment of the extent to which the market acquire commercial characteristics along with examples or instances of information that helps your assessment.

-> We will also support to identify standard/customary terms and conditions such as discounts, warranties, inspection, buyer financing, and acceptance for the Database Encryption industry.

-> We will further help you in finding any price ranges, pricing issues, and determination of price fluctuation of products in Database Encryption industry.

-> Furthermore, we will help you to identify any crucial trends to predict Database Encryption market growth rate up to 2026.

-> Lastly, the analyzed report will predict the general tendency for supply and demand in the Database Encryption market.

Have Any Query? Ask Our Expert @ https://www.verifiedmarketresearch.com/product/global-database-encryption-market-size-and-forecast-to-2025/?utm_source=COD&utm_medium=005

Table of Contents:

Study Coverage: It includes study objectives, years considered for the research study, growth rate and Database Encryption market size of type and application segments, key manufacturers covered, product scope, and highlights of segmental analysis.

Executive Summary: In this section, the report focuses on analysis of macroscopic indicators, market issues, drivers, and trends, competitive landscape, CAGR of the global Database Encryption market, and global production. Under the global production chapter, the authors of the report have included market pricing and trends, global capacity, global production, and global revenue forecasts.

Database Encryption Market Size by Manufacturer: Here, the report concentrates on revenue and production shares of manufacturers for all the years of the forecast period. It also focuses on price by manufacturer and expansion plans and mergers and acquisitions of companies.

About us:

Mr. Edwyne FernandesCall: +1 (650) 781 4080Email: [emailprotected]

Tags: Database Encryption Market Size, Database Encryption Market Trends, Database Encryption Market Growth, Database Encryption Market Forecast, Database Encryption Market Analysis

View post:
Database Encryption Market Segmentation, Application, Technology, Analysis Research Report and Forecast to 2026 - Cole of Duty

Facebook makes it official: WhatsApp audio and video calls will soon have up to 8 participants – GSMArena.com news – GSMArena.com

Posted on April 26, 2020 by admin

The rumors were correct, Facebook is indeed upping the maximum limit of people allowed to participate in a group audio or video call on WhatsApp.

The current limit is 4, but this will "soon" double to 8 for everyone, the company revealed today. The new functionality has already been spotted in beta versions of the app. As before, these calls, whether audio or video, are end-to-end encrypted, meaning no one else can view or listen to them, not even WhatsApp or Facebook.

That's in stark contrast to the also just unveiled Messenger Rooms, which don't come with end-to-end encryption. On the other hand, Rooms does allow for up to 50 people to be on a call.

Unfortunately, Facebook hasn't shared a specific launch timeline for the new maximum limit of participants for WhatsApp. Previous clues found in beta versions of the app implied that all of the 8 participants would have to use the latest iteration of the app, once the functionality launches in the wild. We'll have to wait and see if that pans out.

Source

Follow this link:
Facebook makes it official: WhatsApp audio and video calls will soon have up to 8 participants - GSMArena.com news - GSMArena.com

Microsoft Office 365: How these Azure machine-learning services will make you more productive and efficient – TechRepublic

Posted on April 26, 2020 by admin

Office can now suggest better phrases in Word or entire replies in Outlook, design your PowerPoint slides, and coach you on presenting them. Microsoft built those features with Azure Machine Learning and big models - while keeping your Office 365 data private.

The Microsoft Office clients have been getting smarter for several years: the first version of Editor arrived in Word in 2016, based on Bing's machine learning, and it's now been extended to include the promised Ideas feature with extra capabilities. More and more of the new Office features in the various Microsoft 365 subscriptions are underpinned by machine learning.

You get the basic spelling and grammar checking in any version of Word. But if you have a subscription, Word, Outlook and a new Microsoft Editor browser extension will be able to warn you if you're phrasing something badly, using gendered idioms so common that you may not notice who they exclude, hewing so closely to the way your research sources phrased something that you need to either write it in your own words or enter a citation, or just not sticking to your chosen punctuation rules.

SEE:Choosing your Windows 7 exit strategy: Four options(TechRepublic Premium)

Word can use the real-world number comparisons that Bing has had for a while to make large numbers more comprehensible. It can also translate the acronyms you use inside your organization -- and distinguish them from what someone in another industry would mean by them. It can even recognise that those few words in bold are a heading and ask if you want to switch to a heading style so they show up in the table of contents.

Outlook on iOS uses machine learning to turn the timestamp on an email to a friendlier 'half an hour ago' when you have it read out your messages. Mobile and web Outlook use machine learning and natural-language processing to suggest three quick replies for some messages, which might include scheduling a meeting.

Excel has the same natural-language queries for spreadsheets as Power BI, letting you ask questions about your data. PowerPoint Designer can automatically crop pictures, put them in the right place on the slide and suggest a layout and design; it uses machine learning for text and slide structure analysis, image categorisation, recommending content to include and ranking the layout suggestions it makes. The Presenter Coach tells you if you're slouching, talking in a monotone or staring down at your screen all the time while you're talking, using machine learning to analyse your voice and posture from your webcam.

How PowerPoint Designer uses AML (Azure Machine Learning).

Image: Microsoft

Many of these features are built using the Azure Machine Learning service, Erez Barak, partner group program manager for AI Platform Management, told TechRepublic. At the other extreme, some call the pre-built Azure Cognitive Services APIs for things like speech recognition in the presentation coach, as well as captioning PowerPoint presentations in real-time and live translation into 60-plus languages (and those APIs are themselves built using AML).

Other features are based on customising pre-trained models like Turing Neural Language Generation, a seventeen-billion parameter deep-learning language model that can answer questions, complete sentences and summarize text -- useful for suggesting alternative phrases in Editor or email replies in Outlook. "We use those models in Office after applying some transfer learning to customise them," Barak explained. "We leverage a lot of data, not directly but by the transfer learning we do; that's based on big data to give us a strong natural-language understanding base. For everything we do in Office requires that context; we try to leverage the data we have from big models -- from the Turing model especially given its size and its leadership position in the market -- in order to solve for specific Office problems."

AML is a machine-learning platform for both Microsoft product teams and customers to build intelligent features that can plug into business processes. It provides automated pipelines that take large amounts of data stored in Azure Data Lake, merge and pre-process the raw data, and feed them into distributed training running in parallel across multiple VMs and GPUs. The machine-learning version of the automated deployment common in DevOps is known as MLOps. Office machine-learning models are often built using frameworks like PyTorch or TensorFlow; the PowerPoint team uses a lot of Python and Jupiter notebooks.

The Office data scientists experiment with multiple different models and variations; the best model then gets stored back into Azure Data Lake and downloaded into AML using the ONNX runtime (open-sourced by Microsoft and Facebook) to run in production without having to be rebuilt. "Packaging the models in the ONNX runtime, especially for PowerPoint Designer, helps us to normalise the models, which is great for MLOps; as you tie these into pipelines, the more normalised assets you have, the easier, simpler and more productive that process becomes," said Barak.

ONNX also helps with performance when it comes to running the models in Office, especially for Designer. "If you think about the number of inference calls or scoring calls happening, performance is key: every small percentage and sub-percentage point matters," Barak pointed out.

A tool like Designer that's suggesting background images and videos to use as content needs a lot of compute and GPU to be fast enough. Some of the Turing models are so large that they run on the FPGA-powered Brainwave hardware inside Azure because otherwise they'd be too slow for workloads like answering questions in Bing searches. Office uses the AML compute layer for training and production which, Barak said, "provides normalised access to different types of compute, different types of machines, and also provides a normalised view into the performance of those machines".

"Office's training needs are pretty much bleeding edge: think long-running, GPU-powered, high-bandwidth training jobs that could run for days, sometimes for weeks, across multiple cores, and require a high level of visibility into the end process as well as a high level of reliability," Barak explained. "We leverage a lot of high-performing GPUs for both training the base models and transfer learning." Although the size of training data varies between the scenarios, Barak estimates that fine-tuning the Turing base model with six months of data would use 30-50TB of data (on top of the data used to train the original model).

Acronyms accesses your Office 365 data, because it needs to know which acronyms your organisation uses.

Image: Mary Branscombe/TechRepublic

The data used to train Editor's rewrite suggestions includes documents written by people with dyslexia, and many of the Office AI features use anonymised usage data from Office 365 usage. Acronyms is one of the few features that specifically uses your own Office 365 data, because it needs to find out which acronyms your organisation uses, but that isn't shared with any other Office users. Microsoft also uses public data for many features rather than trying to mine that from private Office documents. The similarity checker uses Bing data, and Editor's sentence rewrite uses public data like Wikipedia as well as public news data to train on.

As the home of so many documents, Office 365 has a wealth of data, but it also has strong compliance policies and processes that Microsoft's data scientists must follow. Those policies change over time as laws change or Office gets accredited to new standards -- "think of it as a moving target of policies and commitments Office has made in the past and will continue to make," Barak suggested. "In order for us to leverage a subset of the Office data in machine learning, naturally, we adhere to all those compliance promises."

LEARN MORE:Office 365 Consumer pricing and features

But models like those used in Presentation Designer need frequent retraining (at least every month) to deal with new data, such as which of the millions of slide designs it suggests get accepted and are retained in presentations. That data is anonymised before it's used for training, and the training is automated with AML pipelines. But it's important to score retrained models consistently with existing models so you can tell when there's an improvement, or if an experiment didn't pan out, so data scientists need repeated access to data.

"People continuously use that, so we continuously have new data around people's preferences and choices, and we want to continuously retrain. We can't have a system that needs to be adjusted over and over again, especially in the world of compliance. We need to have a system that's automatable. That's reproducible -- and frankly, easy enough for those users to use," Barak said.

"They're using AML Data Sets, which allow them to access this data while using the right policies and guard rails, so they're not creating copies of the data -- which is a key piece of keeping the compliance and trust promise we make to customers. Think of them as pointers and views into subsets of the data that data scientists want to use for machine learning."It's not just about access; it's about repeatable access, when the data scientists say 'let's bring in that bigger model, let's do some transfer learning using the data'. It's very dynamic: there's new data because there's more activity or more people [using it]. Then the big models get refreshed on a regular basis. We don't just have one version of the Turing model and then we're done with it; we have continuous versions of that model which we want to put in the hands of data scientists with an end-to-end lifecycle."

Those data sets can be shared without the risk of losing track of the data, which means other data scientists can run experiments on the same data sets. This makes it easier for them to get started developing a new machine-learning model.

Getting AML right for Microsoft product teams also helps enterprises who want to use AML for their own systems. "If we nail the likes and complexities of Office, we enable them to use machine learning in multiple business processes," Barak said. "And at the same time we learn a lot about automation and requirements around compliance that also very much applies to a lot of our third-party customers."

Be your company's Microsoft insider by reading these Windows and Office tips, tricks, and cheat sheets. Delivered Mondays and Wednesdays

Read the rest here:
Microsoft Office 365: How these Azure machine-learning services will make you more productive and efficient - TechRepublic

Apple is on a hiring freeze … except for its Hardware, Machine Learning and AI teams – Thinknum Media

Posted on April 26, 2020 by admin

Word in the tech community is that Apple ($NASDAQ:AAPL) employees are begnning to report hiring freezes for certain groups within the company. But other reports are that hiring is continuing at the Cupertino tech giant. In fact, we've reported on the former.

It turns out that both reports are correct. For some divisions, like Marketing and Corporate Functions, openings have been reduced. But for others, like Hardware and Machine Learning, openings and subsequent hiring appear to be as brisk as ever.

To be clear, overall, job listings at Apple have been cut back.

As recently as mid-March, Apple job listings were nearing the 6,000 mark, which would have been the company's most prolific hiring spree in history. But in late March, it became clear that no one would be going into the office any time soon, and openings quickly began disappearing from Apple's recruitment site. As of this week, openings at Apple are down to 5,240, signaling a decrease in hiring of about 13%.

But not all divisions are stalling their job listings. NeitherApple's "Hardware" or"Machine Learning and AI" groups show a decline in job listings of note.

Hardware openings are flat at worst. Today's 1,570 openings isn't significantly different than a high of 1,600 in March.

Apple's "Machine Learning and AI" group remains as healthy as ever when it comes to new listings being posted to the company's careers sites. As of this week, the team has 334 openings. Last month, that number was 300, an 11% increase in hiring activity.

However, other groups at Apple have seen significant decreases in job listings, including "Software and Services", "Marketing", and "Corporate Functions".

Apple's "Software and Services" team saw a siginificant drop in openings, particularly on April 10, when around 110 openings were cut from the company's recruiting website overnight. Since mid-March, openings on the team have fallen by about 12%.

Between April 14 and April 23, the number of listings for Apple's "Marketing" team dropped by 84. In late March, Apple was seeking 311 people for its Marketing team. Since then, openings have fallen by 36% for the team.

"Corporate Functions" jobs at Apple, which include everything from HR to Finance and Legal, have also seen a steep decline in recent weeks. In late March, Apple listed more than 300 openings for the team. As of this week, it has just around 200 openings, a roughly 1/3 hiring freeze.

So is Apple in the middle of a hiring freeze? Some parts of the company appear frozen. Others appear as hot as ever. Given the in-person nature of Marketing and Corporate Functions jobs, it's not surprising that the company would tap the breaks on interviewing for such positions. On the other hand, engineers working on hardware and machine learning can be remote interviewed and onboarded with equipment delivery.

So, yes, and yes. Apple is, and is not, in the middle of a hiring freeze.

Thinknum tracks companies using the information they post online - jobs, social and web traffic, product sales and app ratings - andcreates data sets that measure factors like hiring, revenue and foot traffic. Data sets may not be fully comprehensive (they only account for what is available on the web), but they can be used to gauge performance factors like staffing and sales.

More:
Apple is on a hiring freeze ... except for its Hardware, Machine Learning and AI teams - Thinknum Media

Machine Learning Market Segmentation, Application, Technology, Analysis Research Report and Forecast to 2026 – Cole of Duty

Posted on April 26, 2020 by admin

H2O.ai and SAS Institute

Global Machine Learning Market Segmentation

To get Incredible Discounts on this Premium Report, Click Here @ https://www.verifiedmarketresearch.com/ask-for-discount/?rid=6487&utm_source=COD&utm_medium=005

Machine Learning Market Region Coverage (Regional Production, Demand & Forecast by Countries etc.):

North America (U.S., Canada, Mexico)

Europe (Germany, U.K., France, Italy, Russia, Spain etc.)

Asia-Pacific (China, India, Japan, Southeast Asia etc.)

South America (Brazil, Argentina etc.)

Middle East & Africa (Saudi Arabia, South Africa etc.)

Some Notable Report Offerings:

-> We will give you an assessment of the extent to which the market acquire commercial characteristics along with examples or instances of information that helps your assessment.

-> We will also support to identify standard/customary terms and conditions such as discounts, warranties, inspection, buyer financing, and acceptance for the Machine Learning industry.

-> We will further help you in finding any price ranges, pricing issues, and determination of price fluctuation of products in Machine Learning industry.

-> Furthermore, we will help you to identify any crucial trends to predict Machine Learning market growth rate up to 2026.

-> Lastly, the analyzed report will predict the general tendency for supply and demand in the Machine Learning market.

Have Any Query? Ask Our Expert @ https://www.verifiedmarketresearch.com/product/global-machine-learning-market-size-and-forecast-to-2026/?utm_source=COD&utm_medium=005

Table of Contents:

Study Coverage: It includes study objectives, years considered for the research study, growth rate and Machine Learning market size of type and application segments, key manufacturers covered, product scope, and highlights of segmental analysis.

Executive Summary: In this section, the report focuses on analysis of macroscopic indicators, market issues, drivers, and trends, competitive landscape, CAGR of the global Machine Learning market, and global production. Under the global production chapter, the authors of the report have included market pricing and trends, global capacity, global production, and global revenue forecasts.

Machine Learning Market Size by Manufacturer: Here, the report concentrates on revenue and production shares of manufacturers for all the years of the forecast period. It also focuses on price by manufacturer and expansion plans and mergers and acquisitions of companies.

About us:

Mr. Edwyne FernandesCall: +1 (650) 781 4080Email: [emailprotected]

Tags: Machine Learning Market Size, Machine Learning Market Trends, Machine Learning Market Growth, Machine Learning Market Forecast, Machine Learning Market Analysis