What’s the best way to secure a world of connected objects?
The world is full of connected devices – and more are coming. In 2017, there were an estimated 8.4 billion internet-enabled thermostats, cameras, streetlights and other electronics. By 2020 that number could exceed 20 billion, and by 2030 there could be 500 billion or more. Because they’ll all be online all the time, each of those devices – whether a voice-recognition personal assistant or a pay-by-phone parking meter or a temperature sensor deep in an industrial robot – will be vulnerable to a cyberattack and could even be part of one.
Today, many “smart” internet-connected devices are made by large companies with well-known brand names, like Google, Apple, Microsoft and Samsung, which have both the technological systems and the marketing incentive to fix any security problems quickly. But that’s not the case in the increasingly crowded world of smaller internet-enabled devices, like light bulbs, doorbells and even packages shipped by UPS. Those devices – and their digital “brains” – are typically made by unknown companies, many in developing countries, without the funds or ability – or the brand-recognition need – to incorporate strong security features.
Insecure “internet of things” devices have already contributed to major cyber-disasters, such as the October 2016 cyberattack on internet routing company Dyn that took down more than 80 popular websites and stalled internet traffic across the U.S. The solution to this problem, in my view as a scholar of “internet of things” technology, blockchain systems and cybersecurity, could be a new way of tracking and distributing security software updates using blockchains.
Today’s big technology companies work hard to keep users safe, but they have set themselves a daunting task: Thousands of complex software packages running on systems all over the world will invariably have errors that make them vulnerable to hackers. They also have teams of researchers and security analysts who try to identify and fix flaws before they cause problems.
When those teams find out about vulnerabilities (whether from their own or others’ work, or from users’ reports of malicious activity), they are well positioned to program updates, and to send them out to users. These companies’ computers, phones and even many software programs connect periodically to their manufacturers’ sites to check for updates, and can download and even install them automatically.
Beyond the staffing needed to track problems and create fixes, that effort requires enormous investment. It requires software to respond to the automated inquiries, storage space for new versions of software, and network bandwidth to send it all out to millions of users quickly. That’s how people’s iPhones, PlayStations and copies of Microsoft Word all stay fairly seamlessly up to date with security fixes.
None of that is happening with the manufacturers of the next generation of internet devices. Take, for example, Hangzhou Xiongmai Technology, based near Shanghai, China. Xiongmai makes internet-connected cameras and accessories under its brand and sells parts to other vendors.
Many of its products – and those of many other similar companies – contained administrative passwords that were set in the factory and were difficult or impossible to change. That left the door open for hackers to connect to Xiongmai-made devices, enter the preset password, take control of webcams or other devices, and generate enormous amounts of malicious internet traffic.
When the problem – and its global scope – became clear, there was little Xiongmai and other manufacturers could do to update their devices. The ability to prevent future cyberattacks like that depends on creating a way these companies can quickly, easily and cheaply issue software updates to customers when flaws are discovered.
Put simply, a blockchain is a transaction-recording computer database that’s stored in many different places at once. In a sense, it’s like a public bulletin board where people can post notices of transactions. Each post must be accompanied by a digital signature, and can never be changed or deleted.
I’m not the only person suggesting using blockchain systems to improve internet-connected devices’ security. In January 2017, a group including U.S. networking giant Cisco, German engineering firm Bosch, Bank of New York Mellon, Chinese electronics maker Foxconn, Dutch cybersecurity company Gemalto and a number of blockchain startup companies formed to develop just such a system.
It would be available for device makers to use in place of creating their own software update infrastructure the way the tech giants have. These smaller companies would have to program their products to check in with a blockchain system periodically to see if there was new software. Then they would securely upload their updates as they developed them. Each device would have a strong cryptographic identity, to ensure the manufacturer is communicating with the right device. As a result, device makers and their customers would know the equipment would efficiently keep its security up to date.
These sorts of systems would have to be easy to program into small devices with limited memory space and processing power. They would need standard ways to communicate and authenticate updates, to tell official messages from hackers’ efforts. Existing blockchains, including Bitcoin SPV and Ethereum Light Client Protocol, look promising. And blockchain innovators will continue to find better ways, making it even easier for billions of “internet of things” devices to check in and update their security automatically.
It will not be enough to develop blockchain-based systems that are capable of protecting “internet of things” devices. If the devices’ manufacturers don’t actually use those systems, everyone’s cybersecurity will still be at risk. Companies that make cheap devices with small profit margins, so they won’t add these layers of protection without help and support from the outside. They’ll need technological assistance and pressure from government regulations and consumer expectations to make the shift from their current practices.