Information Technology

Apple’s upcoming iOS 14.5 release will ship with a feature that will re-route all Safari’s Safe Browsing traffic through Apple-controlled proxy servers as a workaround to preserve user privacy and prevent Google from learning the IP addresses of iOS users.

The new feature, spotted by a Reddit user earlier this week and covered in a report from 8-bit, has been formally confirmed by Maciej Stachowiak, Head of Webkit Engineering at Apple.
The new feature will work only when users activate the “Fraudulent Website Warning” option in the iOS Safari app settings.
This enables support for Google’s Safe Browsing technology in Safari. The Safe Browsing technology works by taking an URL the user is trying to access, sending the URL in an anonymized state to Google’s Safe Browsing servers, where Google accesses the site and scans for threats.
If malware, phishing forms, or other threats are found on the site, Google tells the user’s Safari browser to block access to the site and show a fullscreen red warning.
While years ago, when Google launched the Safe Browsing API, the company knew what sites a user was accessing; in recent years, Google has taken several steps to anonymize data sent from user’s devices via the Safe Browsing feature.
But while Google has anonymized URL strings, by sending the link in a cropped and hashed state, Google still sees the IP address from where a Safe Browsing check comes through.

Apple’s new feature basically takes all these Safe Browsing checks and passes them through an Apple-owned proxy server, making all requests appear as coming from the same IP address.
Many would call the move useless, as Google would still be unable to see what URL the user was checking, but the feature is consistent with other measures Apple has been taking lately, focusing on improving its users’ privacy.
Many of these features have often encroached and disturbed Google’s huge presence in the user analytics and tracking sector.
This includes pioneering broad anti-tracking features in Safari, and forcing app makers to add “privacy labels” to their App Store listings, a requirement that Google has mysteriously avoided by simply not updating any of its apps since last year.
iOS 14.5 is currently in beta and is expected to be released in the coming months. More

Berlin, Germany-based SaaS platform Adjust has released its dating app marketing guide. The guide has benchmarks, spotlights on industry leaders, and has tips on how app developers can retain users by the use of gender targeting, and in-app video streaming.

Over 270 million adults worldwide used dating apps in 2020 and almost two in five (39%) of US adults reported meeting their partner online. However, a major risk to an app’s reputation is the presence of bots on the platform which frustrate the users or exposes them to scams. Fake accounts are generated on a huge scale to engage users and spread spam, link to illicit or explicit sites, or lure people into scams or faker likes to boost specific profiles.
Adjust’s report shows that dating app installs and sessions are at their highest on weekends — Saturdays, in particular. App session length tends to spike early for dating apps, suggesting that users download the app, and quickly match with potential partners.
The buzz and excitement of the app start to drop off toward day 30. Additionally, the report shows that Europeans spend significantly more time in-app than North American or Asian consumers.
Sensor Towers
US dating app downloads have grown to reach a hew high in Q1 2020, despite COVID-19, according to Sensor Towers’ State of Dating apps report, which also shows that younger users are turning to dating apps during the pandemic.
Although the average age for dating apps steadily declined in recent years, the COVID-19 pandemic accelerated this trend in early 2020. The average user age among the top dating apps was around 27.2 years old during the first three quarters of 2020 before jumping back up to 28.2 in Q4. School closures and social distancing orders increased demand among a younger user base.
Unfortunately, dating apps are a hotspot for fake accounts that are trying to scam you. Adjust’s Unbotify feature shows that bots can interact with up to 4,000 profiles within one session. In 2019, the FBI received over 467,000 cybercrime complaints that caused over $35 billion in losses. Approximately 19,473 of those were victims of confidence or romance scams.

So, how can dating app developers reduce the risk for their users? Well, apart from designing the app so that it does not use location-specific advice to track user movements, app developers could include a ‘report’ button so that users can point out when they have interacted with a bot so that it can be eliminated from the platform.
Developers can also use biometric indicators to differentiate bots from real human users and eliminate them from the platform. Users should get smarter about who they interact with and how much information they give away.
You never know, that fabulous potential new partner might be a bot designed to trick you out of your money and break your heart, too. More

British Virgin Islands-based VPN service ExpressVPN Digital Security Lab has released a new report revealing the prominence of location tracker SDKs in dating apps.
Your online privacy is becoming a big deal as apps and business websites track you without your permission and target you due to the information you provide online. But the prominence of questionable location trackers is proliferating among the dating apps you use.
Location data is commonly harvested from your smartphone. It can enrich user-profiles and provide insights into user behavior via intimate details about a user’s movements. Data collected by location and proximity sensors could end up in the hands of law enforcement, intelligence agencies, and military organizations. This massive amount of data about the movements of populations can threaten the privacy of ordinary people around the globe with potential human rights issues.

ExpressVPN Digital Security Lab worked with Esther Onfroy of the Defensive Lab Agency and used the app scanner provided by Exodus Privacy to analyze 450 apps across messaging, gaming, social, and shopping apps used by everyday consumers.
It used a combination of automated tools and manual analysis, to determine whether there are “signatures,” or identifying information, for a tracker in an app’s code, gathering other interesting information such as network endpoints that the app may communicate with.
To do this, it downloaded and unpacked each app installer, disassembled machine language into human-readable source code, searched the source code for tracker signatures and other identifiers, and correlated its findings with web databases, public information, and app stores.
It found that all apps that it analyzed contained questionable trackers. These apps collectively have been downloaded at least 1.7 billion times by consumers globally.

It identified 64 dating apps that have been downloaded at least 52 million times globally. These location trackers are associated with several companies such as X-Mode (subject to a ban by Apple and Google), OneAudience, and Predicio, amongst others, which have repeatedly been called out for privacy violations.

X-mode appeared in 44% (199) of all 450 apps analyzed. Despite the ban, only 10% of these apps have been removed from Google Play.
These dating apps remain available for mass download at the end of January 2021 on the Google Play Store and specifically target a range of sexual orientations and dating preferences, as well as a large assortment of national, ethnic, and racial groups.
These include apps such as Jack’d – Gay Chat & Dating (five million downloads), FEM – Free Lesbian Dating App, Chat and Meet Singles (one million downloads), Encore – Single Parents and Divorced Dating and Chat (500,000 downloads), Black Dating – Meet Online Black Singles Nearby (100,000 downloads), and Asian Mingle – Free Asian Dating and Singles Chat (100,000 downloads).
They also cover more generic dating apps like Mingle2, which claims to have over 39 million members.
There is a growing threat to consumer privacy. When you download an app, you can not take advantage of privacy-protecting searches like Xayn, you are at the mercy of the app. Many apps will not work without location services, and some updates turn settings back on stealthily.
But do you live a life without the apps that bring you joy and keep your location secret, or do you accept that this data may, one day, be used against you in some way? The choice is yours. More

Russian search engine and email provider Yandex said today that it caught one of its employees selling access to user email accounts for personal gains.
The company, which did not disclose the employee’s name, said the person was “one of three system administrators with the necessary access rights to provide technical support” for its Yandex.Mail service.
The Russian company said it’s now in the process of notifying the owners of the 4,887 mailboxes that were compromised and to which the employee sold access to third-parties.
Yandex officials also said they re-secured the compromised accounts and blocked what appeared to be unauthorized logins. They are now asking impacted account owners to change their passwords.
Incident discovered during a routine check
Yandex said it discovered the incident during a “routine screening” by its internal security team but did not elaborate.
The Russian company said that a “thorough internal investigation” of the incident is currently underway and that it plans to make changes to how its administrator staff can access user data.
It also said that there was no evidence to suggest that user payment data was accessed during the recent incident.

While the Russian tech giant said it referred the incident to authorities, a spokesperson did not return a request for comment from ZDNet seeking additional details about the employee and the incident. More

Image: Microsoft
Microsoft says the number of malicious web shells installed on web servers has almost doubled since its last count, last year in August 2020.

In a blog post yesterday, the Redmond company said it detected roughly 140,000 web shells per month between August 2020 and January 2021, up from the 77,000 average it reported last year.
The number has increased as a result of a shift in how hackers view web shells. Once considered a tool for script kiddies defacing websites and the go-to tool of DDoS botnet operators, web shells are now part of the arsenal of ransomware gangs and nation-state hackers alike and are crucial tools used in complex intrusions.
Two of the reasons they have become so popular is their versatility and access they provide to hacked servers.
Web shells, which are nothing more than simple scripts, can be written in almost any programming language that runs on a web server —such as PHP, ASP, JSP, or JS— and such, can be easily hidden inside a website’s source code. This makes detecting them a difficult operation, which often involves a manual analysis from a human operator.
In addition, web shells provide hackers with a simple way to execute commands on a hacked server via a graphical or command-line interface, providing attackers with a simple way to escalate attacks.
Web shells more prevalent as more servers are put online
As the corporate IT space has moved towards hybrid cloud environments, the number of companies running web servers has increased over the past few years, and, in many cases, public-facing servers often have direct connections to internal networks.

As Microsoft’s stats have shown, attackers appear to have figured out this change in the makeup of corporate IT networks as well, and have amped up their attacks on public-facing systems.
Web shells now play a crucial role in their attacks, providing a way to control the hacked server and then orchestrate a pivot to a target’s internal network.
These types of attacks are exactly what the US National Security Agency warned about in April 2020 when it published a list of 25 vulnerabilities that were often used to install web shells.
The NSA report didn’t just warn about web shells used on public-facing systems but also about their use inside internal networks, where they’re used as proxies to jump to non-public-facing systems.
Microsoft urges companies to re-prioritize their approach to dealing with web shells, which are slowly becoming one of today’s biggest security threat. As ways to keep networks secure, the OS maker recommends a few basic actions:
Patch public-facing systems, as most web shells are installed after attackers exploit unpatched vulnerabilities.
Extend antivirus protections to web servers, not just employee workstations.
Network segmentation to limit the damage of an infected server to a small array of systems and not the entire network.
Audit and review logs from web servers frequently, especially for public-facing systems, which are more vulnerable to scans and attacks.
Practice good credential hygiene. Limit the use of accounts with local or domain admin level privileges.
Check your perimeter firewall and proxy to restrict unnecessary access to services, including access to services through non-standard ports. More

Australia’s Consumer Data Right (CDR) officially launched on July 1 with the first tranche, an open banking-like regime, requiring financial services providers to share a customers’ data when requested by the customer.
While the first tranche of the CDR applies to the financial services industry, energy and telecommunications will soon join the regime.
Read more: Australia’s Consumer Data Right: Here’s everything you need to know
Data can only be shared with accredited data recipients (ADRs). But of concern to Australian Information and Privacy Commissioner Angelene Falk is that “big tech” has the ability to apply for ADR status.
“It’s currently open to large technology and social media companies to apply to be accredited as data recipients in the CDR scheme, however, I understand that none are currently accredited and I’m not aware of any specific use cases as to why they may wish to engage, so in a sense, I’m speaking in the abstract,” she said.
Appearing before the Senate Select Committee on Financial Technology and Regulatory Technology on Friday, Falk said one of the strong protections in the CDR system is consumer consent and the ability for individuals to exercise choice and control about how their data is handled.
She’s concerned that this may also give the technology giants access to more data than they already have.

“I think because of the rich data holdings that are held by some of the social media platforms, care would need to be taken to ensure that individuals understand what they’re consenting to if their Consumer Data Right information were to be combined with that [which is] perhaps is on their social media profile,” Falk said.
“Some of the risks I think are around the insights that could be derived from that information and it could include sensitive information and be used in ways that individuals might not expect.”
See also: NZ Privacy Commissioner labels Facebook as ‘morally bankrupt pathological liars’
She pondered whether the committee consider that a digital platform should have access to all data, or whether there be a condition that it not be combined with sensitive data the organisation may already hold.
“There’s other issues around the use of algorithms and artificial intelligence in the combining of data that may lack transparency for consumers and be difficult to explain … [they are] some of the challenges with having fully informed and freely given consent when you enter into very complex data handling arrangements,” she added.
Individuals have the ability to make a complaint if they feel that their personal information has not been handled in accordance with the legislative requirements, and the OAIC has had 20 “contacts” in relation to the CDR system.
“We have a triaging role so that consumers who are engaging in the system don’t need to navigate government in order to make a complaint or make an inquiry, so they’ll come to our office and we’ll triage them to the appropriate entity,” deputy commissioner Elizabeth Hampton explained.
She said of those 20 contacts, the OAIC has had two complaints and eight inquiries for its office; and nine inquiries and one “report” that have been sent to the ACCC.
While those numbers are low, Falk said they reflect the number of people engaged in the system, expecting the number to grow alongside scheme uptake.
MORE FROM THE OAIC More

Brazil’s National Data Protection Authority (ANPD, in the Portuguese acronym) has informed today (11) that it has started an investigation into the country’s second largest data leak of the year.
The investigation relates to the exposure of data relating to more than 102 million mobile phone lines from two mobile operators, which, according to Brazilian news website Neofeed, included names, taxpayer registration numbers, minutes spent on phone calls and other details, including information relating to president Jair Bolsonaro.

A cybercriminal based outside Brazil who claimed to have obtained 57,2 million customer data sets from Vivo and 45,6 million data sets relating to Claro customers has been selling the data in the dark web, the article said. Cybersecurity and privacy firm Psafe discovered the incident on February 3, but couldn’t find evidence that both mobile operators had actually been the source of the leaks – and both companies deny that any customer data has been leaked.
The data protection authority stated that “it is taking all the appropriate measures” to investigate the case. The ANPD has summoned the Federal Police, as well as “the company that reported the fact and the companies involved”. The idea is that the organizations will help the newly-formed authority, which has released its initial strategy last week, to assist in the investigation and the adoption of actions to contain and mitigate risks related to the personal data of the consumers that have potentially been affected.
The news of the latest leak follow a previous incident earlier this year whereby details of 223 million Brazilians, including deceased citizens, ranging from name, address to current income, personal vehicle information and tax returns were exposed and sold in the dark web. More

Security software specialists Cloudflare this afternoon reported Q4 revenue and profit that topped analysts’ expectations, and forecast this quarter’s sales, and the full year, higher as well, citing health in sales to large enterprises. 
Despite the upbeat report, Cloudflare shares dropped 7% in late trading.
Chief executive and co-founder Matthew Prince called it “a remarkable end to a year we’ll never forget.”
Prince noted the company had “delivered more than 550 products and capabilities during 2020 that also supported needs bigger than all of us—whether it was helping to secure the US election from cyberattacks or ensuring COVID-19 vaccine registration sites withstand demand with Project Fair Shot.
Cloudflare’s revenue in the three months ended in December rose 50%, year over year, to $125.9 million, yielding a net loss of 2 cents a share, excluding some costs.
Analysts had been modeling $118 million in revenue and a 4-cent loss per share.
Cloudflare said it had a dollar-based net retention rate in the quarter of 119%, up 3 percentage points from the prior quarter, which it said was “driven by continued strength from large enterprise customers.”

For the current quarter, the company expects revenue in a range of $130 million to $131 million, and a net loss of 2 cents to 3 cents a share. That compares to consensus for $126 million and a 3-cent loss. 
For the full year, Cloudflare projects revenue of $589 million to $593 million, and a net loss of 8 cents to 9 cents per share. That compares to consensus for $561 million and 9 cents. 

Tech Earnings More