Why we are building Synnax, and why Credit Intelligence is the future

As many of you know, I have a deep interest in corporate financial analysis. However, there are inherent limitations to how much a single person can analyze, and it takes significant time and expertise to catch the nuances of increasingly complex financial reports. Combined with an ever-evolving market and macroeconomic environment, it becomes evident that mastering the skills of a proficient credit analyst is time-consuming, especially when timely decisions are crucial.

Despite advancements like the Internet and blockchain technology, credit analysis has struggled to keep pace with the rapid evolution of financial markets. The core issue is straightforward: processes relying on human input are challenging to scale. Additionally, human biases often influence analyses, leading to decisions that prioritize confirming preconceived notions over objective evaluation of all available data. This bias is particularly perilous in credit analysis, where errors can significantly impact crucial decisions. The article “Credit Scores and the Bias Behind Them” provides pertinent examples of this issue.

With the advent of blockchain technology offering unparalleled transparency and reliability of data, and the capability to run artificial intelligence (AI) models at scale and cost-effectively, it is time to move beyond outdated methods. For the past 50 years, credit analysis has not evolved significantly, with the primary change being the shift from paper to Excel spreadsheets. This is why we are building Synnax.

A Case Study: The Collapse of Three Arrows Capital (3AC)

Anyone involved in the Web3 space is familiar with the downfall of Three Arrows Capital (3AC). If you need a refresher, this link provides a quick recap. The critical issue was that almost everyone, including their lending counterparts and trading partners, was blindsided by the events as they unfolded.

How could Synnax have predicted this and mitigated the damage to 3AC’s stakeholders?

1. Data-Driven Predictions: Synnax operates on a decentralized network of data scientists developing machine learning models to predict a company’s financial strength in real time. These models focus solely on data, ignoring subjective “trust me” documents. In 3AC’s case, the models would have quickly detected discrepancies between the real-time value of assets and liabilities and the net asset value (NAV) publicly declared by the company.
2. Real-Time Intelligence: Through Synnax’s granular credit intelligence and forward-looking probability predictions from decentralized ML models, lenders and trading partners would have had real-time data to question the figures shared by 3AC. For example, 3AC’s significant position in the GBTC arbitrage trade involved borrowing funds to buy Bitcoin, tendering them to Greyscale, and receiving GBTC units traded at a premium in traditional markets. They then used these units as collateral to borrow more funds. However, when the arbitrage opportunity diminished, the spread collapsed, and 3AC’s positions began to deteriorate. The ML models would have identified the increasing risk and deviations in 3AC’s NAV early on.
3. Early Warning Signals: As 3AC’s GBTC positions went underwater, the Synnax protocol would have provided early warning signals to lenders through predictive outputs. This would have allowed lenders to reassess their exposure before the situation worsened.
4. Unbiased Analysis: Unlike traditional credit analysis, Synnax’s ML models are free from human biases, ensuring objective assessment based solely on data. This would have revealed the true extent of 3AC’s financial troubles much earlier.
5. Transparency and Reliability: Even if 3AC had withheld data, Synnax’s Zero-Knowledge Proof of Equity (ZK/e) score would have indicated a low transparency score, affecting the default probability calculations. A lower ZK/e score would have resulted in a higher probability of default metric, alerting stakeholders to the increased risk.

Synnax’s approach is equally effective for traditional companies (Web2) because the models prioritize data over subjective human judgment, mitigating biases and improving risk assessment accuracy.

By leveraging blockchain and AI, Synnax aims to revolutionize credit analysis, providing a more transparent, reliable, and scalable solution to address the complexities of modern financial markets.

Finally, one thing I have not addressed in this article, is that all of the above can be done in a privacy-preserving way. Synnax also leverages cutting-edge encryption and secure computation technologies to protect data privacy. We will get into the weeds of this in a future article.