The NIST AI Risk Management Framework provides a governance, risk, and compliance framework for artificial intelligence.
The document highlights the challenges in AI risk management, including identifying and cataloging risks, emergent risks, and availability of reliable metrics.
The criteria to evaluate AI systems include validity, safety, security, accountability, transparency, privacy, and fairness in managing harmful bias.
Bank runs can still happen today, triggered by various factors like social media rumors and liquidity issues
The FDIC, established in 1933, guarantees a certain amount of depositors' funds to prevent bank runs and provide stability to the banking system
Modern banking faces new challenges like swift fund transfers, online activism affecting stock prices, and the need for regulators to adapt to the changing landscape
Modularization is key in project management. Breaking down a large project into repeatable modules simplifies delivery.
Different projects have distinct distribution patterns, impacting their likelihood of success. Thin-tailed distribution projects are more likely to be on time and budget.
Lessons from the book can be applied to personal projects like kitchen renovation or holiday planning. It's a valuable read for project managers and those tackling significant endeavors.
AIS vessel tracking data can predict China's exports, monitor global trade, and understand real-time economic activity.
Data cleansing is crucial for turning raw AIS data into actionable insights. Cleaning the data involves filtering out anomalies and ensuring accuracy.
It's important to consider limitations like the exclusive focus on large commercial ships, uncertainties in cargo data, and vessel behavior anomalies when analyzing AIS data.
Institutional investors need to find surprising insights in data but also be skeptical of them to ensure accuracy and avoid errors.
When using alternative data to make predictions, it's crucial to verify if the insights answer the right questions and differ from the market consensus.
Digging into the data through various methods like independent validation, error margin assessment, and data integrity checks is essential for investors to ensure the reliability of surprising insights.
Credit Suisse faced major financial issues over several years, with losses, leaked data, and fines leading to its demise.
Swiss authorities forced Credit Suisse to merge with UBS to prevent a public collapse and global financial crisis.
The fallout from Credit Suisse's fall includes impacts on European banks, market confidence, and central bank efforts to stabilize the financial system.
The purpose of stress tests for banks in peacetime is not necessarily to predict future crises, but to ensure banks have enough capital and that the tests are tough and variable.
It's important for stress test scenarios to change and remain tough to prevent banks from manipulating their capital levels and misrepresenting their financial health.
The public stress test process during peacetime may not have a significant impact on capital allocation to the banking sector, unlike crisis-time stress tests.
Silicon Valley Bank faced a significant spike in insider loans, raising concerns about VC conflicts of interest and Director independence.
Approximately 38% of SVB's reported incremental venture debt went to start-ups affiliated with Directors, prompting questions on underwriting transparency.
The lack of disclosure in the Proxy Statement and the concentration of loans to insider-affiliated start-ups suggest potential VC conflicts of interest impacting risk management.
After the collapse of Silicon Valley Bank, the American banking industry is adapting by relying less on uninsured deposits and more on alternative funding methods like borrowings.
Deposits have restabilized post-SVB crisis, but banks are facing challenges with tight lending standards due to renewed economic pessimism and liquidity concerns.
Banks are cautiously navigating post-SVB crisis by reducing reliance on uninsured deposits, managing securities losses, and addressing liquidity worries amid tighter monetary policy.
Buying the dip strategy involves waiting for an asset price to drop below a specific threshold before purchasing it, but simulation data shows that this strategy is usually less effective than buying regularly.
When dealing with volatile assets, buying the dip can be beneficial if the asset underperforms in the median case but significantly overperforms occasionally, providing exposure without heavy losses.
For stable assets or normal investors, buying regularly is usually the best strategy as it requires less effort and is generally more effective than trying to time the market by waiting for price dips.
Venture capital industry can exhibit herd behavior and be consensus-driven, deviating from contrarian thinking.
Investors chasing beta during market peaks can lead to significant losses when the market corrects, highlighting the importance of sound investment strategies.
Sticking to core investment sectors, avoiding chasing trends, and focusing on long-term conviction can lead to sustainable returns independent of market cycles.