Stock Markets July 10, 2026 10:09 AM

JPMorgan Tests AI Agents That Outperformed 60/40 in Historical Backtests

Bank strategists report AI-driven allocation systems topped a traditional stock-bond mix in simulations, but warn results are preliminary

By Caleb Monroe
Share
Twitter Reddit Facebook LinkedIn
JPM

Shares of JPMorgan rose modestly after the bank disclosed that internally developed AI-driven investing agents outperformed a conventional 60/40 stock-bond portfolio in historical simulations covering roughly two decades. The top-performing agent delivered about 0.7 percentage point of annual outperformance with lower volatility versus the 60/40 benchmark, and also bested the bank's rules-based market regime model. Strategists cautioned the findings come from backtests, not live trading, and urged careful integration of agentic AI into asset allocation processes.

JPMorgan Tests AI Agents That Outperformed 60/40 in Historical Backtests
JPM
Summarize with
ChatGPT Perplexity Claude Grok Gemini

Key Points

  • JPMorgan disclosed that internally developed AI investing agents outperformed a traditional 60/40 stock-bond portfolio in historical simulations covering about two decades.
  • The top AI system delivered roughly 0.7 percentage point of annual outperformance and showed lower volatility versus the 60/40 benchmark, while also beating the bank's rules-based market regime model.
  • Results are based on backtests, not live trading, and strategists stress that agentic AI must be embedded within thoughtful asset allocation processes.

JPMorgan shares ticked higher on Friday after the bank disclosed that AI-powered investing agents beat a classic 60/40 portfolio in historical simulations. The note revealed systems designed to rotate between equities and fixed income in response to evolving market conditions.

In backtests spanning the past two decades, the highest-performing AI system outpaced a 60/40 mix by 0.7 percentage point per year while registering lower volatility, the bank's strategists said. The same system also outperformed the firm's rules-based market regime model, according to the strategists led by Thomas Salopek.

Those results come from historical simulations rather than real-world trading. The strategists warned against treating the backtests as definitive evidence that AI can reliably beat markets in live conditions. "The AI agent can be set up with a process to be empowered to make decisions under uncertainty, producing outperformance vs a reasonable benchmark," the strategists reportedly wrote Thursday, describing the project as the firm's initial attempt to construct an AI system for identifying market regimes.

The strategists also flagged important caveats around wider adoption of such approaches. "We strongly caution against uncritically accepting what amounts to in-sample, overly confident answers of AI," they wrote. "Agentic AI needs to be grounded in a well thought-out asset allocation process, rather than naively assuming the agent can be the source of the domain knowledge."

Bank researchers built the agents to shift exposures between stocks and bonds based on signals tied to changing market environments. The work represents an exploratory effort to see whether machine-driven decision agents can improve capital allocation across markets.

Financial institutions have spent the past two years incorporating large language models and related systems into research, coding and internal investing tools. The bank's note indicates those efforts are evolving into trials that test whether such systems can take on capital allocation roles - albeit under careful oversight and with explicit recognition of limitations from backtested results.


Context and next steps

The bank described this as an initial foray rather than a finished product. Strategists emphasized the need to embed agentic AI within established allocation frameworks and to guard against overconfidence in in-sample performance. The findings will likely shape further internal testing rather than immediate changes to client-facing products.

Risks

  • The findings are derived from historical simulations rather than live investing, so past backtested performance may not translate into future results - this affects investment management and wealth management sectors.
  • Overreliance on in-sample AI outputs could produce overly confident decisions if the models are not grounded in robust allocation frameworks - this is a risk for institutional asset allocators and portfolio managers.
  • Widespread adoption of agentic AI without proper guardrails could lead to operational or decision-making risks across trading and research functions within banks and investment firms.

More from Stock Markets

JPMorgan Publishes 24 Top Short Ideas for Q3 2026 Across Eight Sectors Jul 10, 2026 Moroccan Market Edges Up as Banking, Beverage and Transport Stocks Lead Gains Jul 10, 2026 Asahi’s June Performance Stalls as Weather and Promotions Weigh on Japan Beer Sales Jul 10, 2026 Shopify orders merchants to delist e-cigarettes from stores Jul 10, 2026 Oslo OBX closes lower as media, transport and diversified financials weigh on market Jul 10, 2026