In 2015, The Economists describes the blockchain technology as "a machine for creating trust," whose "extraordinary potential" is overlooked due to the shady image of bitcoin. International Trust Machines Corporation (ITM) was founded four years later, in the January of 2019, to offer blockchain-enabled chips as a solution to scalability problem of public ledgers.
The next decade will see 75 billion devices connected to each other and interact in various manners, as statistics reveals. The ensuing explosion of data volume, with IoT applications alone producing 90 zettabytes (90 billion terabytes) annually by 2025, leads to substantial business opportunities. But as the numbers of transactions and participants increase in the network, the demand for scalability should be met to be able to handle it.
ITM, after enabling blockchain to scale up, wants the data from IoT devices to be recorded once they're connected to the Internet.
Building trust in data
CEO and co-founder Julian Chen believes that immutable data on the blockchain are valuable in this era when fraud has been rampant in every industry.
In a distributed ledger, it's possible to build mutual trust in data with consensus mechanism. Rather than a third party, the data will be submitted to a blockchain under peer-to-peer protocols.
Despite numerous advantages of blockchain, it's difficult to leverage the technology for IoT. Co-founder and chief scientist Gwan-Hwan Hwang pointed out three common bottlenecks to overcome:
The blockchain system throughput is measured in the number of TPS (transactions per second) it's capable of supporting. Most of the public chains accommodate much less transactions than those that centralized systems do. Also, since in blockchain, nodes need to reach consensus before recording the data, it will be impossible for a blockchain to process the endless flood of data from a large group of IoT devices. It's a trade-off between trust and speed.
For example, GDPR, an EU law on data protection and privacy, entitles users to the right to request an erasure of any personal data, but it's not possible on blockchain, a platform on which all stored data are immutable. There's also a dilemma in the trust mechanism when it comes to recording certain types of transactions like real estate transactions.
Miners are paid based on the sets of data they record. It would cost an exorbitant amount of money if all the data from their IoT devices go to a blockchain.
Focusing on the convergence of IoT and blockchain, ITM implements a cryptographic security algorithm in the chips, what Taiwan is well known for, to allow anchoring and M2M (machine to machine) auditing. Under an exclusive security protocol, cryptographic calculations will be performed on the hardware level before the transactions are recorded in the blockchain.
Within their framework, a package of transaction data on the local database (e.g. on an IoT device) will be given a fingerprint to be recorded in the public blockchain, where the data accuracy will be verified. A single transaction on the public chain therefore can clear millions sets of off-chain data while ensuring that the data is trustworthy and tamper-proof.
ITM applies the solution to three major scenarios: battery management, smart meter, and smart transportation.
To speed up its growth, the team is actively looking for partners. It has formed an alliance of blockchain-enabled IC providers with Taiwan's Institute for Information Industry (III) and cooperated with fabless IC designer MyTek. "Once the transactions are recorded as soon as a device gets connected to the Internet, there will be more possibilities for 'trusted data as a service' as a business model," Chen said.