Tag

hackers

Browsing

By Jennifer Jolly

Here’s how to combat targeted ads, hackers, and cyber thieves.

Sometimes I feel like I spend my entire day dodging attacks on my digital identity. I wake up to scammy text messages begging me to click sketchy links. I delete them. Brush my teeth.

Open my email and voila! Fake sweepstakes emails — from my own email address, no less — telling me I won everything from a power drill to a Yeti cooler. I delete those, too. By noon I’ve silenced at least a half dozen robocalls, and at least once a day I see a Facebook ad for something I recently talked to my husband about — is Siri eavesdropping on me, too?

Screengrab of scam email from me.
Screengrab of scam email from me.
Jennifer Jolly

Obviously, I’m not alone. With so many scams floating around we’re all starting to see privacy dangers around every mouse click, even where they might not exist, like in a Snapchat filter.

The recent midterm elections and upcoming Black Friday/Cyber Monday online shopping extravaganza have only made these concerns more intense. If you haven’t gotten at least a hundred unsolicited text messages — again, with sketchy-looking links — consider yourself lucky.

Where is all of this headed? Are we forever doomed to a future of digital paranoia, and the threat of cybercrime, stolen money, identities or worse? Is there a way to break free from all these shady spammers, scammers and thieves?

The good news is: It all gets a lot less scary once you realize what is going on.

Now you know: 7 default settings tech companies don’t ever want you to change

An email from myself?

I get emails from myself all the time … only I never sent them. They’re often low-effort scam-bait messages claiming that I won something or I have unclaimed funds somewhere. Or even more annoying, that I’ve been hacked “watching porn” on my laptop and better pay up — or else.

Spoiler alert, there’s no watching porn or getting hacked actually going on, these are among the most common of threats.

You likely get these, too. No doubt, looking at your own email address in the “From” line is unsettling, but how does that even happen?

Sadly, it’s easy. The email addresses that populate when you open an email are rarely verified, especially if you use a free email service. Using a less-secure Simple Mail Transfer Protocol (SMTP) server, a scammer can just type in what they want the “To” and “From” addresses to say.

Someone using one of these servers — a scammer can just set one up themselves — can make an email look like it came from anyone, including you. It really is that easy.

As annoying as it is to change an email service, you can avoid these spoofed emails and do away with a whole lot of spam and scam messages by switching to a secure email provider.

ProtonMail is a popular one these days. It’s free for a private email account, and since it uses more advanced protocols than most free services, it spots spoofed emails, so you don’t have to.

Online scams targeting veterans and active duty members. Here’s how to protect yourself.

Password Paranoia

Oh, passwords. I don’t know of a single person on the planet that actually likes the password system, but there’s no getting around them … or is there?

“Passwords have been the default mechanism for authentication since the beginning of computer technology,” Bob Eckel, president and CEO of Aware, a biometrics software company based out of Massachusetts, tells me over email.

“They don’t require extra or special hardware, there are no compatibility issues and they are a cost-effective option for companies and businesses of all sizes, which is why they are still the go-to for identity or use authentication today.”

The biggest problem with passwords is that they’re far less secure now than they were a couple of decades ago. Modern hackers use more advanced techniques, and most of us don’t practice good password hygiene, such as using a different password for every account.

If you use your birthday, maiden name, pet’s name or one of the most easily hacked passwords on the planet, like 1234, or “password,” you’re just begging to get scammed.

Companies like Apple and Google pioneered new methods for securing digital devices such as smartphones and even laptops, but individual accounts for the millions of apps we all use still require passwords. Our own fear and apprehension are a big part of the reason that certain biometric technologies aren’t catching on as fast as they could.

RNC sues Google over email spam filters: Alleges ‘bias against Republicans’ ahead of midterms

“Certain segments of the general public, such as baby boomers, for example, continue to be weary of facial authentication; instead, they’re much more receptive to fingerprints. Therefore, we need to continue our mission to educate both organizations and consumers about the technology and special techniques used to make facial authentication highly safe, secure, and accurate,” Eckel says.

While we wait for passwords to kick the bucket, utilizing features like Apple’s “Sign in with Apple” can effectively bypass many app login requirements and use your smartphone as the default for verification.

It’s also a lot more convenient than inventing a new password for every app, and lets you use FaceID or TouchID (depending on your device) to log into just about anything.

Man who lied about being a millionaire: Fake Navy SEAL stole up to $1.5M in romance scam, DOJ says

Is Siri listening?

Do you ever see an ad for something you were just talking about with a friend, within earshot of your phone?

It’s easy to assume that since our phones are constantly listening to us — waiting for the trigger word, like “Hey Siri,” “Okay Google,” or even “Alexa” — that they may be working behind the scenes to feed us relevant ads. That’s not exactly right, but it’s not that far off either.

“Our phones are designed to listen, first and foremost, to virtually assist us, which can explain why you may be served ads that directly relate to a conversation you just had,” Eckel adds. “It is similar to how search engines work by tracking your activity to ensure it is delivering the most relevant results.”

But it’s important to note that Apple, Amazon and Google have all stated that they treat the audio from their respective virtual assistants with the utmost security and privacy.

Tab overload? How to control what happens when you start your browser

You may get a relevant ad on your computer, related to a voice prompt if you choose to search Amazon’s marketplace using Alexa, for example, but having a background conversation with a friend isn’t the reason you get those ads.

The more likely scenario is that your searches on desktop or mobile triggered an ad algorithm to suggest those relevant products for you.

I know it can seem creepy and weird, but unless the big three companies are lying through their teeth — and security researchers haven’t busted them yet — it’s just a coincidence and a cleverly-designed ad system.

It’s not you, it’s them: Google, Alexa and Siri may answer even if you haven’t called

Steps to be more secure online

Duckduckgo screenshot.
Duckduckgo screenshot.
Jennifer Jolly

What can you do about it all? Here are some simple steps you can take to search for information online as privately as possible these days:

  • Use a privacy-focused search engine that doesn’t collect as much data about your habits as say, Google. Options here include DuckDuckGo or Brave Search.
  • Disable your mobile ad ID on your smartphone and tablet, and block ads on your laptop and desktop browsers. Most smartphone apps default to collecting tons of data about your behavior. There are easy ways to disable these functions in the settings on iPhones and Androids that run on the Google operating system.
  • Use an ad blocker such as Ghostery or AdBlock Plus.

By Jennifer Jolly

Jennifer Jolly is an Emmy Award-winning consumer tech columnist. Email her at j[email protected]. Follow her on Twitter: @JenniferJolly. The views and opinions expressed in this column are the author’s and do not necessarily reflect those of USA TODAY.

Sourced from US Today Tech

Sourced from WIRED.

Meet the exciting individuals and groups bringing about the next wave of digital security in the second phase of the Re-Inventors series highlights.

From your fridge, your oven and your toothbrush to your car, doorbell and bike lock, more and more everyday objects are being imbued with smart features and internet connectivity. All of which means the physical world around us is becoming more hackable. WIRED and HP showcase ten re-inventors working to keep this rapidly growing number of vulnerable devices safe from cyberattack.

Mischa Dohler (above)

Professor in wireless communications, King’s College London

On October 21, 2016, a massive distributed denial of service (DDoS) attack brought down the sites of Twitter, Pinterest, Reddit, Spotify, PayPal, Verizon, Comcast, and a host of other major US web companies. Behind it: your smart fridge, home security camera or DVR, all of which were lassoed via the publicly available malware, Mirai, into a vast botnet and directed at major domain name system service provider, DYN.

The attack would have come as no surprise to King’s College London Professor in Wireless Communications, Mischa Dohler, who recently cautioned that the vulnerabilities of unobtrusive and unchecked Internet of Things (IoT) devices make the rise of such attacks only a matter of time.

For the Mirai malware, enlisting devices came easily, thanks to out-of-date firmware and the prevalence of default admin logins, allowing it to build a DDoS army numbering, according to DYN, in the tens of millions. Unless the industry develops new cybersecurity solutions fast, this, Dohler has emphasised, is only the beginning.

Ovidiu Vermesan

Co-ordinator, IoT European Research Cluster (IERC)

For Ovidiu Vermesan, whose role is to facilitate collaborative research on how the Internet of Things can support European development, the potential of IoT is enormous. From smart sensors for a more efficient National Grid, and congestion-aware road signs to responsive city lighting and self-regulating buildings – in sufficient quantity tiny devices can have a large impact. First, however, they need to be secure. The vulnerability of your thermostat is one thing, but as similarly small devices are connected up to home oil rigs, and motorway junctions, the need to protect against cyber attack is becoming ever more critical

Samu Konttinen

Executive vice president, Consumer Security, F-Secure

The average antivirus program runs to a few hundred megabytes in size. The average IoT device has no more than a couple of gigabytes of memory in total. Take firmware and data storage into account and that leaves little space for security software. Instead Helinski-based F-Secure are taking security back a step, to the router itself. Its Sense hub scans the network on behalf of all the devices connected to it, from lightweight systems like connected security cameras to the traditional heavy-weight desktop PC.

“People are starting to buy all kinds of new smart devices to use in their homes, and we know that these devices are already being hacked,” said Samu Konttinen, F-Secure executive vice president, Consumer Security when announcing SENSE. “We’ve seen footage from nanny cams streamed online without people’s knowledge or consent, and it’s been proven that intruders can use something as simple as a connected light bulb to get access to people’s homes.”

But security doesn’t end at home. F-Secure also runs another level of security checks from the cloud, allowing a lightweight client on the device to query incoming connections from anywhere in the world before accepting them. “Criminals are constantly developing new ways of turning people’s technology against them,” Konttinen said. “Security solutions need to evolve to meet this challenge.”

Carlos Moreira

Founder and CEO, WISeKey

The number of connected IoT devices is expected to hit 6.4 billion this year, according to US research firm, Gartner, a 30 per cent rise from 2015. By 2020 the size of the IoT network is expected to reach 20.8 billion things. Keeping each device safe is enough of a challenge, but there’s also a need to keep your network safe from each device and the potential of allowing access to even just one malicious connection among millions.

That’s where WISeKey’s Root of Trust comes in. Incorporated into the hardware of the chip itself this provides a cryptographic signature of the device’s authenticity and integrity to an artificially intelligent cloud-based security platform that learns from attack attempts.

“This new platform opens up many possibilities for improving the overall security functionalities by taking advantage of the world of digital identification at the object level,” explained Carlos Moreira, founder and CEO WISeKey. ”Once combined with big-data this allows us to gather and process all available information to generate actionable insights.”

Patricia Lewis

Research director, International Security, Chatham House

For Patricia Lewis, one of the greatest cybersecurity threats are the devices almost all of us rely on, and few of us ever see. Satellites, the subject of a research paper Lewis co-authored in September, underpin everything from navigation and air traffic control to financial services, defence and weather forecasting. They also cost enormous sums of money to build and launch, leading to minimal turnover and significant legacy design issues, including vulnerability to cyberattack.

This is not only a potential concern for the future. Several times since 2012, North Korea has jammed GPS signals to South Korea, taking Seoul’s mobile network down for several days at a time. More dangerous is the potential for spoofing attacks which send false signals instead, potentially causing ships to collide or bringing down vital financial and energy infrastructures. Alternatively, by taking control of the satellite itself, hackers could crash it into other satellites or bring it crashing back down to Earth.

“Technology alone cannot provide the basis for policymaking on cybersecurity,” the report concluded. What is needed, Lewis recommended, is, “An international ‘community of the willing’ – made up of able states and other critical stakeholders within the international space supply chain… to develop a space cybersecurity regime competent to match the range of threats.”

Boris Danev (above)

Founder and CEO, 3DB Technologies

Convenience often comes at a security cost, as Boris Danev knows all too well. In 2010, while a Phd student at ETH Zurich, Danev discovered the increasingly popular passive keyless entry systems – which use Bluetooth to allow an owner to unlock and start their car by proximity alone – could be very easily hacked. All he needed was a cheap off-the-shelf Bluetooth amplifier to trick the car into thinking the key was much closer – even when it was actually up to 165 feet away on the dining room table inside the car owner’s house. Some insurers now refuse to cover such cars unless additional security, such as a locked garage, is in place.

Through Thalwil, his Switzerland-based startup, 3DB Technologies, Danev is now commercialising a chip specially designed to prevent such attacks. This works by sending very short energy pulses at a specific frequency range, just like a password, which changes on every use. The key then returns the signal back to the car, which uses the elapsed time to measure the exact distance between them.

Neil Costigan

CEO, BehavioSec

Imagine a future without passwords. Neil Costigan does, but that’s not to say he’s looking to a future without security. Instead Costigan believes we can replace a several character code with something both much more secure, and much easier to remember – your behaviour. Specifically your unique behavioural interaction patterns from the way you touch your phone, or move your mouse, or type words on a keyboard. Spun out from research conducted at Luleå University of Technology in 2009, BehavioSec’s system has been shown to detect an intrusion within milliseconds with a 99.7 per cent accuracy. Over 2015, they provided security for 1.2 billion transactions for 15 million consumer users and have since expanded to Germany, the UK, the US and Benelux.

Sebastien Marcel

Senior research scientist, Idiap Research

From facial features and speech habits, to the pattern of veins in our fingers, behaviour is far from the only natural unique identifier that human’s posses as a potential password replacement. But when introducing any new security system the most important task is figuring out the ways it could be fooled. At the Switzerland-based Idiap Research Institute, Sebastien Marcel leads the Biometrics group in researching how multiple authentication modalities can be combined for more secure authentication solutions, how a unique biometric identifier can be spoofed, and then how to prevent such spoofing from being successful.

Dudu Mimran

CTO, Telekom Innovation Laboratories, Ben Gurion University

Your smartphone, according to Israel-based Dudu Mimran, is increasingly becoming a hacker’s number one target. Nothing is better suited to the capitalisation of a user slip-up, from opening a malware-containing file to entering your security details to a false website than a device we carry around almost all day.

The challenges in securing it, however, are many. The roll-out of security fixes is slow across the fragmented Android ecosystem, and we rely on thousands of micro-business app-developers to keep their own piece of the security puzzle up to date. A further complication Mimran points to is the mixing of business and private uses in the same device, allowing any malicious software to enter through a novelty app downloaded for personal use, then to gain access to everything from your fitness records to your enterprise CRM app.

Ofer Ben-Noon (above)

Co-founder and CEO at Argus Cyber Security

Having your hard drive wiped by malware may be infuriating, but as our cars are become internet connected devices the cybersecurity stakes are becoming even higher. This was made starkly obvious by two security researchers at the Black Hat security conference, who showcased the ability to remotely paralyse a Jeep Cherokee’s engine, or disable the brakes at low speeds.

Fortunately, at Tel Aviv-based Argus Cyber Security, Ofer Ben-Noon has led the development of a solution to keep our cars safe. Argus’ Intrusion Prevention System (IPS), embedded during production, uses cloud-based algorithms to monitor every signal sent across the car’s network to remotely detect and block suspicious transmissions, while the ability to remotely update the system prevents the need for a costly physical recall.

Learn more about how HP is helping to re-invent security.

Sourced from WIRED