Technical FAQs
Last updated
Last updated
Blockchain and distributed ledger technology enables the construction of a secure and privacy-preserving data infrastructure. Our solution provides personal data security and privacy using the basics of blockchain - public key cryptography - whilst enabling the data to be analysed - using compute-to-data - and therefore useful to the user.
A compute-to-data model provides a means to exchange data while preserving data privacy by allowing data consumers to run compute jobs on the data (which can be stored on local servers, on existing cloud providers, or on a decentralised file system such as IPFS) to train AI models. Rather than having the data sent to where the algorithm runs, the algorithm runs where the data is.
In traditional IT environments, personal data is stored and managed by organisations in such a way that individuals have no ability to control it. However, using distributed ledger technologies (such as blockchain), users own their data and this is referenced by a data token, which can only be accessed by the owner's private keys.
The consensus mechanism used in blockchain is more energy-intensive than current centralised data storage solutions. The proof-of-work mechanism (used to verify each Bitcoin transaction) is the most energy-intensive consensus mechanism as miners compete against each other with computing power to mine a block.
Other consensus mechanisms exist, for example, proof-of-stake or proof-of-authority, which use a fraction of this energy (~99.95% less). Instead of all nodes competing to mine each block, a miner (or group of miners) are chosen at random to resolve each block. Ethereum is moving towards a proof-of-stake mechanism, whilst Layer 2 solutions such as Polygon are already using proof-of-stake.
Furthermore, other technologies such as the IOTA’s Tangle, a directed acyclic graph, ensure every participant is contributing to the validation process of the transactions, so that consensus is reached without the need for dedicated miners. Since this validation does not require any heavy computations, the network can be maintained by relatively small devices, with limited power and calculation capacity.
