JPMorgan tests AI investment agent capable of autonomously adjusting stock and bond allocations

BlockBeats News, July 11, JPMorgan is testing AI agents capable of autonomously adjusting the allocation between stocks and bonds in order to dynamically rebalance portfolios based on changes in market conditions.

Test results show that, in 20 years of historical backtesting, the AI model with the best performance achieved an annualized return 0.7 percentage points higher than the traditional “60/40” stock-bond portfolio, while also exhibiting lower volatility. All 8 AI agents tested by JPMorgan delivered higher risk-adjusted returns.

However, JPMorgan stated that the results are still based on simulated tests and not actual investment performance. The bank also warned that large-scale adoption of AI could lead to convergence of trading strategies, increase crowded trading in the market, and amplify market volatility under stressful conditions.