Three individuals are presenting in front of a large screen displaying the Signet platform.

Home
Student Projects
Signet: Cryptographic Compliance Rails For Agentic Finance On Xrpl

Signet: Cryptographic compliance rails for agentic finance on XRPL

Ripple – UBRI Blockchain XRPL & Policy Gateway Project

Our project was Signet: XRPL guardrails for agents transacting in the financial sector. Essentially, it's a way for AI Agents to be deployed safely for purposes such as asset management and trading by adding a layer of security with human escalation that ensures the system can only do actions it’s intended for.

"We’re getting to the point where AI Agents are going to have an unprecedented impact: millions of dollars and even human lives at stake. This is precisely why the moment for enforcing guardrails is now". (Ricardo).

Purpose of the project

What was the main purpose behind choosing this project, and what impact or real world problem were you aiming to address?

The main reason behind choosing this project was that as agents are being adopted by the general public, they are prone to be involved in high-stakes industries such as healthcare and financial service, where mistakes are very costly. We wanted to solve this problem, starting with the financial sector.

Technical Details

What technical approaches and tools did you use, and why did you choose them?

We developed an MVP using the XRPL and some of its primitives. XRPL is a blockchain that looks to bridge the gap between industries such as banking and Web3, thus, it was very natural for us to choose it as part of our stack. Furthermore, XRPL offers some primitives related to the multi-signature of transactions that allow us to have a policy (a set of rules) and a human (if escalated) to sign a transaction for it to be approved. This was the fundamental technology we applied to solve the problem. Furthermore, we used some additional properties of XRPL, such as enclaves (a secure, off-chain trusted execution environment), to ensure that the policy cannot be jailbroken by bad actors.

What knowledge, tools, or theories from your program courses did you apply to this project?

Most of the basics we obtained from the XRPL Course, however, some of us had also taken Blockchain before as a part of our degree in Computer Science and AI.

Challenges and Solutions

Were there any significant challenges during this experience, and how did you overcome them? Did what you learned in the program help you along the way?

The main challenges we faced were associated with coming up with a solid, trust-worthy way to communicate our solution. Blockchain and Crypto are very controversial subjects, and agentic AI, though less controversial, is a very popular topic nowadays. Thus, it was very important for us to identify a way to communicate our intentions and our solution in the right way to ensure that it was received with the intention we wanted it to.

Collaboration and Teamwork

How did teamwork contribute to the success or progress of your project?

Teamwork was super important! This started in Paris, when we were seated on a table just talking about the future and coming up with ideas. Moments like this sparked what today is Signet, and we believe that collaboration and team building is crucial to ensure that these ideas come to life. In terms of implementation, we were also lucky to be a group of different individuals with a similar background in relation to Computer Science. This allowed us to iterate quickly and ship Signet successfully.

Learning and Takeaways

What key lessons and skills (soft and hard) have you acquired or strengthened through this experience?

Collaboration: We are better at collaborating and improving on our ideas than we were before. As said earlier, most of our ideas and progress came from casual conversations where each of us imparted a bit of what they knew, and collaborating properly was crucial.
Organization: Developing a working product in a short period of time is not easy. We had to delegate, prioritize, and work efficiently, which greatly improved our organization skills.
Communication: We also learned how to communicate an idea that may be controversial/scary to some audiences in an effective manner.
Blockchain/XRPL: Before this project, we had surface knowledge of Blockchain and Cryptocurrencies, but we did not have a clue of how XRP worked. Now, we are capable of implementing products on the ecosystem, and probably, would be able to do so as well in other blockchains.

Future Development

Do you have plans for further development or improvement of your project in the future?

We do plan on continuing to work on it, however, we do not have a clear timeline or plan to do so. Some of us would like to be on the research side, while others are interested in pursuing the product side.

ADVICE

What advice would you give to other students who want to get involved in this type of initiatives and tech immersions?

I would say that the most important thing right now is taking a leap and beginning. AI is making complex technologies significantly more accessible to everyone, thus, it is important to take this opportunity to do things we find interesting, try them out, and pivot if they fail. But the core advice I would give is just opening a Claude session, brainstorming, reading and learning about the problem you want to solve, and getting something going.