Decentralized confidential computing, better known as DeCC for simplicity’s sake, is under the spotlight in Messari’s latest industry report. The blockchain research firm has published an extensive report that highlights the $1B in investment that’s been sunk into DeCC projects in recent years and which is now bearing fruit.
Messari gives particular scrutiny to Garbled Circuits, the confidential computing technology best known for its application on COTI’s Layer 2 network. It provides a primer on how Garbled Circuits works and outlines some of its most powerful use cases, before following up with other leading DeCC technologies such as MPC and ZKPs. Here are three main takeaways from Messari’s wide-ranging report.
1. DeCC Is in High Demand
While some onchain narratives appear to be driven by speculation or pure hopium, DeCC has a compelling use case. It’s not about forcing a tech for which there’s little organic demand, but rather about meeting a need, particularly among enterprises, for greater privacy controls when interacting onchain. Industries such as healthcare, banking, and education could get a lot out of blockchain, given its ability to support open collaboration between organizations, but there needs to be privacy protections in place to safeguard data.
This is what DeCC brings to the table, allowing businesses to enjoy the best of both worlds: the openness of blockchain coupled with the data privacy they’re accustomed to. This is about more than simply enabling private onchain transactions, however: DeCC allows data to be computed onchain without revealing its content. This enables smart contracts to execute based on a specific event – say a record query matching a patient’s SQL database entry – without broadcasting that information publicly.
2. Garbled Circuits Has a Lot Going For It
Messari devotes a lot of its report to analyzing COTI’s implementation of Garbled Circuits, which enables secure computation by allowing one party to evaluate a function on encrypted data without learning the input. This is achieved by encrypting both the data and the function to ensure privacy during computation.
Messari notes that “COTI supports a wide range of real-world use cases, from consumer payments to enterprise integrations and government-grade digital currency pilots. For highly regulated environments, COTI’s proprietary Data Privacy Framework (DPF) allows regulatory audits of encrypted data without compromising user privacy, ideal for regulated sectors such as banking, lending, and government.”
It also highlights the ways in which Garbled Circuits outperforms comparable DeCC solutions, emphasizing the fact that it’s 1,000x faster and 250x more storage-efficient than FHE-based alternatives. On COTI’s L2, the report touches upon current benchmarks that show “80–100 confidential transactions per second (ctps) for ERC-20 operations, with potential for ~1000 ctps with upcoming optimizations.”
3. ZKPs Continue to Impress With Their Versatility
There’s a lot that can be done with zero-knowledge proofs (ZKPs), particularly in the context of confidential computing, where their privacy-preserving capabilities prove their worth. Penumbra, Aleph Zero, and Aleo are cited as some of the web3 projects working with ZKPs in this regard. For example, Aleo’s L1 blockchain supports fully private computation, allowing developers to create applications where sensitive user data and logic remain confidential. This is achieved using the chain’s Leo programming language, which compiles to ZK circuits.
In assessing the capabilities of all the DeCC solutions that feature in its report, Messari notes “the difficulty of preserving confidentiality in systems designed for openness.” What all of these projects are broadly trying to achieve is privacy in the background with a normal user experience in the foreground. This is harder than it sounds, since encrypted data is more computationally intensive to process. But thanks to breakthroughs in confidential computing, from MPC to Garbled Circuits, these challenges are being chipped away at.
If DeCC can achieve its full potential, tomorrow’s dapps won’t just be capable of handling exponentially more data than before: they’ll do so while preserving the privacy of enterprise and retail users alike. With AI agents also starting to get up to speed, and demands for vast amounts of data to feed the AI-web3 juggernaut intensifying, it’s not just businesses that stand to benefit from confidential computing: so do the machines whose intelligence is closely correlated to the quality and quantity of data they’re fed.
