Earlier today, we hosted an AMA about the COTI V2 white paper.
The AMA took place on the COTI official Telegram group. If you weren’t able to join, here is the full recap:
Natalie:
Hey everyone, thank you for joining us today for our special COTI V2 White Paper AMA! My name is Natalie, COTI’s Marketing Manager, and I will be your host today.
Joining me today are Shahaf Bar-Geffen, COTI’s CEO, Dr. Nir Haloani, COTI’s CTO, Yair Lavi, COTI’s Chief Strategy Officer and Dr. Avishay Yanai, Soda-Labs CEO. Welcome everyone!
The AMA will consist of two parts: in the first part we’ll ask pre-submitted questions and in the second part, we will open the conversation up to everyone. So without further adieu, let’s begin.
We are going to start with the community pre-submitted questions.
For those who are not very technical, what are the most important things to understand from the WP?
Shahaf:
I’ll take this one please.
In the whitepaper, we introduced gcEVM: An innovative extension of the Ethereum Virtual Machine (EVM) designed by Soda Labs to infuse confidentiality into blockchain transactions. gcEVM allows for secure computations on encrypted data, ensuring data privacy without compromising functionality. gcEVM unlocks the full potential of Ethereum for sectors that demand high confidentiality and security standards.
We also introduced a novel concept to our L2 design: Decentralized Sequencers. The Decentralized Sequencers design approach aims to address the inherent limitations of centralized sequencers — such as single points of failure and censorship risks — by distributing the sequencing function across multiple participants. This approach marks a leap toward a more robust, efficient, fair and democratic blockchain ecosystem, distinguishing COTI V2 from other Layer 2 solutions reliant on centralized sequencing mechanisms.
Natalie:
Thank you, Shahaf! Next question: Isn’t privacy against regulations?
Shahaf:
As Chris Dixon of a16z puts it: It’s a popular misconception that blockchains enable secrecy and anonymity. In fact, Blockchains are so public by default that their innate transparency could actually slow adoption. People may not use blockchains if they fear doing so will expose sensitive information, such as salaries, medical bills, or payment history.
Or as I put it: Sensitive data on a public blockchain is a bug, not a feature.
So no, Privacy is not against regulation. On the contrary, confidentiality is a regulatory requirement in a lot of circumstances. To be compliant, many organizations need to be able to prove that they not only protect user information from data breaches, but also make fair and appropriate use of the data. Think GDPR, Hippa etc.
Unlike anonymity, confidentiality deals with the protection of specific personal or private information. It explains what information can be viewed and who has permission to access it.
Natalie:
Great answer, thanks! Can you explain what Garbling Circuits are?
Shahaf:
Garbling circuits are unique because they allow you to use encrypted data, do an encrypted calculation on it, and return an encrypted value that only the recipient can decipher.
In very simple terms, imagine a six-piece puzzle with each piece being owned by a single individual. For whatever reason, these six people don’t trust each other, and no-one is willing to discuss which piece they own. Garbled circuits allow a mechanism where all six people can work together to complete the puzzle, without revealing their identity or which piece they initially held.
This is what’s called end-to-end encryption and because data never needs to be deciphered, it is the ultimate protection of sensitive information. Any data leak that has ever occurred would have been avoided if data was encrypted the entire time. GC does that and it’s compatible with the encryption standards out there.
Avishay:
Garbled circuit (GC) is not only an encryption of data, but an encryption of functions; meaning that using GC can make any classic symmetric key encryption scheme, in particular those used by governments and large banks, to have homomorphic properties — so you can do computation on them.
The technology is around from about 1982, the security of garbling schemes are fully proven in the literature in the most rigorous approach.
Natalie:
That was a great answer. Next, we have this question for Dr. Haloani:
What are decentralized Sequencers and how does COTI V2 plan to maintain decentralization and prevent centralization of network control over time?
Dr. Haloani:
A sequencer in a blockchain network is responsible for ordering transactions before they are finalized on the blockchain. Centralized sequencers, while efficient, pose risks including single points of failure and potential censorship.
Decentralizing the sequencer function distributes these responsibilities across multiple participants, thereby enhancing security, redundancy, and resistance to censorship. In COTI we introduce a Shared Sequencer Network: A cluster of sequencers that operate in a rollup-agnostic manner and ensure the network’s liveliness and decentralization. Decentralized Sequencers bring a novel concept to our L2 design.
Natalie:
Thank you Nir. Could you tell us what’s the main difference between fully homomorphic encryption (FHE) and garbled circuits?
Shahaf:
FHE does end-to-end encryption too but requires all those involved to change their well established encryption techniques to FHE, which is expensive, tasking and dangerous. You could say that GC allows you to turn standard encryption techniques to FHE. Also, in terms of comparison to other techniques, GC are about 1000x faster and less expensive in terms of computation, 250X better in terms of latency and 100X lighter in storage.
It was said that FHE is the holy grail of encryption. I’ll tell you what I think: GC is a holier grail!
Development is easy as it’s the same EVM, only it’s called gcEVM and you can just choose if data or an operator is private.
Natalie:
Thank you Shahaf. The next question is also for you: COTI mentioned AI as one of the use cases which is a front running narrative this cycle. How does COTI plan to implement AI and do you plan on promoting this more since it’s such a massive narrative?
Shahaf:
COTI plans to unlock confidential machine learning by introducing GCML.
COTI V2 allows for an entirely new paradigm in which AI / Machine Learning models can be safely and effectively trained on sensitive datasets that cannot be traced back to an individual or business. COTI V2 will use PPML (privacy-preserving machine learning) through cryptographic techniques like garbled circuits (GC); these models could do so in a way that doesn’t violate the privacy of those whose data it uses, opening up whole new business models.
Natalie:
That sounds amazing! What about wallets? How will they use COTI V2 and is there a plan for a new wallet?
Shahaf:
We see confidential TX as one of our key use cases, enabling mass adoption and serving as a bridge to the traditional world. Wallets are integral to this. COTI V2 will enable parties to transact in a manner that conceals the specific amount transferred while maintaining traceable confidentiality, which can be tailored to comply with regulatory standards.
We have a new wallet partnership planned.
Natalie:
And what about Defi? How can Defi leverage COTI V2?
Shahaf:
As we mentioned before, we are taking all the best parts of Ethereum DeFi and adding to it confidentiality by choice.
For example, confidential smart contracts that can be tailored to obscure transaction data while leaving the address data of involved parties visible. It is practically institutional-grade DeFi that meets data privacy regulations.
This brings many possibilities for native MEV protection and novel DeFi primitives, such as:
Private AMMs — enabling automated market makers that offer confidentiality for trading pairs, safeguarding sensitive trading information while ensuring transparency of participant addresses.
Dark pools- Dark pools can be implemented using COTI V2, allowing for off-exchange trading venues where order book data is hidden, ensuring privacy for large trades while maintaining the integrity of participant identities.
under-collateralized lending: Under-collateralized lending can be facilitated through COTI V2’s confidential smart contracts, enabling lending protocols where the specifics of loan terms and borrower identities are concealed, enhancing privacy while mitigating risk.
Blind auctions: Blind auctions can be conducted on COTI V2, where bid amounts and bidder identities are kept confidential until the auction concludes, ensuring fairness and privacy for participants.
Hybrid DEXs — combining elements of centralized and decentralized exchanges while incorporating confidentiality features, offering traders a blend of liquidity, security, and privacy in their trading activities.
In fact, I’m in Paris Blockchain week right now and my next meeting is with a Defi giant.
Natalie:
That’s really awesome!
Thanks for the elaborated answer. We have so much to look forward to! Now we have a more technical question. Dr. Haloani, can you tell us which framework will the L2 network utilize? Using Cosmos sdk, Optimism OP stack etc
Dr. Haloani:
We haven’t made a final decision yet, but we’re currently considering both options. For the decentralized sequencer, we’re looking into several platforms, including Espresso, Astria, Radius, and AltLayer.
Natalie:
Thanks for the answer. Next up, we have this question for Avishay:
Can I ask if there is a patent granted for the whole technology with the GC?
Avishay:
Yes, there are multiple pending patents.
Natalie:
Here’s our next question: In COTI v2 it’s stated that there will be no need for a bridge. How will the tokens enter & exit between the L2 and the eth mainnet? Will that happen via a smart contract?
Yair:
As a matter of fact, the bridge will still be active and can be used to move between ETH and L2, however since L2 native token is EVM, it will be supported by ETH wallets and most exchanges. The bridge will be completely rebuilt so the issues we see today will not exist
With that in mind, we’re aware of the current bridge issues being raised by the community, we’re working on fixing them in the next few days and once it’s done we will double the flow on the bridge, allowing transactions of 25,000 $COTI per user per day (instead of 10,000)
Natalie:
That’s great news! Thanks for your answer, Yair. Next, we have the following question:
Will current Viper wallet holders have to KYC all over again with Civic, or are we all going to be grandfathered over?
Shahaf:
I think you mixed up the info a bit.
For COTI V2, KYC will not be mandatory on the protocol level like it is today. dApps developers and specific payment solutions can determine if KYC is required for their specific to be compliant.
Civic will utilize COTI V2’s dynamic decentralized identity solution to guarantee complete data protection and confidentiality for its users. This means that Civic’s users who possess a digital identity will be able to access a dynamic digital identity and utilize it across various dApps with COTI’s V2 confidential layer.
Natalie:
Thanks for clarifying that.
We have 1 more pre-submitted question, and then we will open the chat for community questions. Can you elaborate on sequencers for node operators such as will there be tiered staking, rewards and if the current cohort will still get priority?
Shahaf:
Specific details about validators in COTI V2 will be shared at a later stage, but we will prioritize the current testnet and mainnet node operators.
Natalie:
That’s great news!
Thanks guys! Now let’s open the chat for some live questions. Once there are many questions we will mute the chat again to give the team time to answer. Let’s go! 🔥
Community:
Do you have any plans for the gCoti airdrop campaign again in the near future?
Shahaf:
Use the treasury to get gCOTI for free + grow your rewards for the V2 airdrop.
Community:
What happens to the DAG based CMD tokens once Coti V2 is live?
Shahaf:
For a while both networks will coexist, but over time you’ll swap to the V2 native version.
Community:
1) About Shared decentralized Sequencers. Does this mean that a dApp made on Polygon can leverage COTI’s privacy capabilities without switching the whole dapp on COTI?
2) About FHE vs Garbled circuits and your claims that GC is 1000x faster. As you said GC was developed many years ago but before your research couldn’t be deployed on blockchain. As I’ve read, the same goes for FHE. Zama and Fhenix state that they have made breakthroughs in FHE that now makes it usable on blockchain too. Is your GV vs FHE comparison on speed based on “old” FHE schemes? Do you think Zama may have found a way to speed it up too like you did with GC?
3) To my understanding GC is destroying competition in the privacy sector. Is there any application that you think FHE would be actually a better fit than GC?
Dr. Haloani:
1) Yes, a dApp built on Polygon could potentially leverage COTI’s privacy features without needing to migrate entirely to COTI, thanks to the shared decentralized sequencers and interoperability protocols outlined in the COTI V2 Whitepape
Avishay:
2) The claim is based on benchmarks published by zama, not old FHE schemes (FHE doesn’t exist so long anyway).
3) In web3 privacy solutions will be based on MPC. GC and FHE are two ways to achieve MPC, and in that space there is no use case that is fitted to FHE more than GC. But there are use cases fitted more for GC than FHE, due to performance issues.
Community:
Hello, Shahaf! I support COTI for a long time, nice job done and good luck!!
My question is: As far as i could understand main COTI v1 features were implementing infrastructure on DAG, that allowed to process ~100 000 transactions per second. Will this solution be changed or not after transition on eth l2?
Shahaf:
We will be using decentralized sequencers over Ethereum instead
Community:
Wow, COTI v2 is a fundamentally different approach!
Do you see the future of DJED project?
Shahaf:
Djed is running on Cardano so it’s really independent. We have great plans for Djed
Community:
Will the current exchanges support the native token of COTI on the L2 network when v2 launches? Or only the one on the eth network?
Yair:
We expect most major exchanges that support COTI will support V2
Community:
What will be the role of existing V1 nodes after V2
Shahaf:
For a while both will co-exist and then it’s all V2. We prioritize those who were node operators and testers in V1 for the future roles that V2 has. There are more roles.
Community:
Are you attending token 2049 next week?
Shahaf:
Yes I am
Community:
I understand that in order for GC to work, users encrypt their data on their potentially low performance devices using already established cryptographic schemes and send them to COTI network to be processed without decryption. For this processing on encrypted data, nodes on COTI have to be extremely powerful with special GPUs and CPUs?
Avishay:
Nope, In the long run you’ll be able to participate from your phone
Community:
Can GC & decentralized sequencers be done with other blockchains and is the patent only with ETH.
Shahaf:
Not right now, but I hope it’ll become the market standard in the future, as that will be the key for adoption
Community:
Can we ask questions about your vision of current cryptomarket situation and prospects of COTI in it?
Shahaf:
maybe I can do it in my next video
Community:
Any update about PPN?
Shahaf:
As mentioned in my video, we’re slowly scaling up with our client. Will keep you posted.
Shahaf:
ok folks, I’m off to conquer the world in Paris blockchain week. Things look good for us!
Natalie:
Thank you everyone for these excellent questions, and thank you Shahaf, Nir, Yair and Avishay for answering them!
That is all the time we have for today.
For those who haven’t had the chance to read the COTI V2 white paper, you can find it here: https://coti.io/files/coti_v2_whitepaper.pdf
Have a great day everyone!
For all of our updates and to join the conversation, be sure to check out our channels:
Website: https://coti.io/
X: https://twitter.com/COTInetwork
Telegram: https://t.me/COTInetwork
GitHub: https://github.com/coti-io
Discord: https://discord.gg/9tq6CP6XrT
