IoT global round-up: Mirai embraces capitalism as botnets boom

14 Apr 20178 Shares

Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Pin on PinterestShare on RedditEmail this to someone

Botnets such as Mirai rely on overwhelming numbers for attacks. Image: BeeBright/Shutterstock

Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Pin on PinterestShare on RedditEmail this to someone

In a week full of funding and trucks, it was the Mirai botnet and a push for the internet of trees that caught the eye.

Start-ups in industrial internet of things, machine learning, virtual reality (VR) and ‘machine linking’ enjoyed funding windfalls this week as more financiers eyed a future, connected reality.

Branch, for example, secured $60m to help press ahead with its software that allows developers to provide better analytics and understanding of how websites are referred to by apps. Virtualitics landed $4.4m in funding to provide analytics for VR and augmented reality applications.

Mirai IoT

Deep into its funding journey, LPWA (low-power wide-area) networks operation Actility raised $75m in a Series D round this week. Elsewhere, Elmodis, a start-up looking at machine communications to protect against any dangerous malfunctions, raised $4.9m to bring its total funding past the $5m mark.

And the money talk stretches beyond clever start-ups, with Apple rumoured to be interested in Toshiba, teaming up with Taiwanese contract manufacturer Foxconn to acquire a 20pc stake in Toshiba’s semiconductor business.

Toshiba is the world’s second-biggest flash memory chipmaker, but has put its chips business up for sale to make up for a write-down of $6.5bn in its US nuclear equipment operations.

In connected vehicle news, Tesla revealed its ambition to look past the sale of saloons and SUVs, and take on the commercial sector. According to the BBC, Tesla will reveal a pickup truck – called the Model P – within the next 18 to 24 months.

Here’s what else you may have missed in the world of internet of things (IoT):

Mirai eyes a bit of money

Mirai is the name of the botnet that silenced US infosec journalist Brian Krebs, of Krebs on Security, in a 620Gbps DDOS attack late last year.

The same botnet took French hosting provider OVH offline after enlisting around 145,000 IoT devices to hack CCTV cameras.

It replicated this on an even larger scale when millions of US internet users lost access to popular online sites such as Twitter, Spotify, Reddit and CNN. At the time, it was considered one of the largest take-downs in internet history.

Well, Mirai didn’t go away. Nothing ever does, online. Instead, it became open source and a newer version has been detected. As well as being able to issue DDoS attacks and the like, it is now equipped to mine bitcoin.

According to IBM, a new variant called the ELF Linux/Mirai malware is designed to scan for devices running Telnet services and compromise them.

The infected nodes are then used to perform further attacks. ELF Linux/Mirai is targeting DVR (dvrHelper), web IP cameras on BusyBox, and other BusyBox-powered Linux IoT boxes as well as unattended Linux servers.

“There are also DDoS and bitcoin-miner components built into this malware,” said IBM. No longer a random attack, Mirai is going corporate.

Linux attacks grow

Continuing on a similar theme, another Linux-targeted attack emerged from a piece of malware called BrickerBot. Used in a similar way to Mirai, BrickerBot attacks unsecured devices that have not changed the default username and password provided on purchase.

According to ReadWrite, BrickerBot distances itself from Mirai in one key way, though. Once inside the unsecured device, it starts to permanently remove the storage and revokes internet access, effectively killing the unit.

“This is the major difference between Mirai and BrickerBot; while Mirai uses the corrupt IoT devices, BrickerBot makes them unusable,” ReadWrite reports.

Radware, the cybersecurity company that found BrickerBot, put forward a few points of advice for those wary of an attack affecting their IoT devices:

  • Change the device’s factory default credentials.
  • Disable Telnet access to the device.
  • Network behavioural analysis can detect anomalies in traffic and combine with automatic signature generation for protection.
  • User/entity behavioural analysis (UEBA) can spot granular anomalies in traffic early.
  • An IPS should block Telnet default credentials or reset Telnet connections. Use a signature to detect the provided command sequences.

Internet of trees

This week, Coillte, Ireland’s national forestry company, revealed a €1m deal with Treemetrics to develop a real-time satellite communications and data analytics platform for forestry harvest control.

Treemetrics is the company behind the ‘internet of trees’ plan, connecting forestry stakeholders to better understand, monitor and manage woodlands.

GPS and various mobile comms operators will help to build an Irish element to what will one day become a ‘world supply chain’ of forestry produce.

“Coillte is on a path to becoming the best forestry and land solutions company in Europe,” said Fergal Leamy, chief executive of the company.

“Our increasing use of technology will be crucial over the next few years as we aim to create the connected forest. I am delighted to be working with world-leading technology companies like Treemetrics, who also happen to be based in Ireland, as we develop innovative solutions to how we manage our forests.”

AI spots racial and gender biases

Researchers have made a rather troubling discovery, with machines proving to be reflections of their creators in surprising ways.

The researchers found that common machine learning programs, when trained with ordinary human language available online, can acquire cultural biases embedded in the patterns of wording.

These biases range from the morally neutral, such as a preference for flowers over insects, to the objectionable views of race and gender.

“Questions about fairness and bias in machine learning are tremendously important for our society,” said researcher Arvind Narayanan, assistant professor at Princeton University.

“We have a situation where these artificial intelligence systems may be perpetuating historical patterns of bias that we might find socially unacceptable and which we might be trying to move away from.”

Want stories like this and more direct to your inbox? Sign up for Tech Trends, Silicon Republic’s weekly digest of need-to-know tech news.

Gordon Hunt is a journalist at Siliconrepublic.com

editorial@siliconrepublic.com