Is the evolving set of artificial intelligence (AI) platforms based on large language models interesting as retirement portfolio specification advisors? Are they monolithic, or diverse? As a simple exploration, we pose to each of Grok, ChatGPT, Claude, Perplexity and Gemini the following prompt regarding exchange-traded fund (ETF) selections and weights:
For a hypothetical U.S. investor with moderate risk tolerance at each of ages 30, 40, 50, 60, 70 and 80, please provide your unique view on the three ETFs, with annually rebalanced fixed weights, that the investor should hold in a retirement portfolio over the next 10 years. Do not provide any explanation.
We then compare and contrast results from AI panel members. Using responses to the prompt as posed in early June 2026, we find that:
Read this study free
The rest of this study: 110 words and 1 chart.
Enter your email and we’ll send you a link to this article. 3 premium articles a month, no password or card.
Are Target Retirement Date Funds Attractive? PremiumVanguard Target Retirement 2020-2050 funds were generally not preferable to a monthly rebalanced 60% SPY/40% LQD mix on return, volatility and reward-to-risk metrics (inceptions…
Testing the All Weather Portfolio PremiumThe annually rebalanced 60/40 VTI/TLT benchmark beat the All Weather portfolio (30% VTI, 40% TLT, 15% IEF, 7.5% GLD, 7.5% DBC; annual rebalance) in…
Review of the Golden Butterfly Portfolio PremiumGolden Butterfly (equal-weight VTI/IJS/TLT/SHY/GLD; monthly or annual rebalancing) had markedly lower CAGR than VTI and GLD but a fairly shallow maximum drawdown (ETFs Nov…
Best Defense for a 60/40 Portfolio? PremiumDefensive Absolute Return (DAR; monthly long lowest-/short highest-correlation thirds of 25 long/short factors) averaged 0.8% gross monthly return in the 60/40 portfolio's worst 10%…
No Safe Fixed Retirement Withdrawal Rate? PremiumWith frictions included, the inflation-adjusted 4% withdrawal rule failed for 1960s retirees across 100% stocks, 60/40, 30/70 and 100% corporate bond allocations, but worked…