Investing Expertise

What lessons are emerging from individual investor use of artificial intelligence (AI) platforms based on large language models (LLM) as an aid? As a simple exploration, we ask each of Grok, ChatGPT, Claude, Perplexity and Gemini to search for examples of individuals using LLMs to guide investments and provide a very concise set of lessons learned from their experiences. Using responses from these AIs to this request as posed in mid-October 2026, we find that: Keep Reading

How should investors view financial advice from artificial intelligence (AI) models based on large language models, which have neither finance credentials nor investing track records? In their October 2026 paper entitled “Judge for Yourself: Confidence Beats Accuracy in AI Financial Advice”, Golnaz Bahrami, Anthony Cookson and Austin Moss study advice preferences on Chatbot Arena, where users vote on answers to same question from two randomly assigned and anonymous AI models. Model anonymity guarantees that users do not vote based on AI reputation. Votes may indicate a decisive preference between the two answers, or they may result in a tie (both answers good or bad). For each AI answer, they task Claude Sonnet 5 to: (1) grade its accuracy based on Certified Financial Planner Board practice standards and life cycle prescriptions; and, (2) assess AI confidence in and actionability of the answer. Using votes on 3,060 financial questions (2,212 of which are decisive) from two public releases of Chatbot Arena full conversations during June to August 2024 and April to July 2025 and Claude answer assessments, they find that:

Keep Reading

Does the evolving set of artificial intelligence (AI) platforms based on large language models offer a quick way to evaluate an active exchange-traded fund (ETF) that “seeks to offer a distinct source of monthly income by selling a 30-barrier put option?” As a simple exploration, we pose to each of Grok, ChatGPT, Claude, Perplexity and Gemini the following prompt:

Please review the attached documents and recommend whether an investor seeking income with typical risk aversion should allocate to Simplify Barrier Income ETF (SBAR) rather than some other ETF.

The attached documents are: (1) the SBAR Summary Prospectus of November 2025; (2) the SBAR Fact Sheet 3Q26; and, (3) the full SBAR price history. Compiling and comparing responses from the AI panel to the prompt as posed in late October 2026, we find that: Keep Reading

In their October 2026 paper entitled “When Price Charts Rhyme: Historical Visual Retrieval for Stock Return Prediction”, Hui Chen, Haoyu Liu, Ke Wu and Dexin Zhou report gross annualized hedge portfolio returns as high as 85% and gross annualized Sharpe ratios as high as 6.37 from strategies that predict returns by matching current price patterns to past patterns/subsequent returns. Their sample runs from 1993 through 2025. Their hedge portfolio is long (short) the tenth, or decile, of stocks with the highest (lowest) predicted returns, reformed weekly. Can an artificial intelligence (AI) model quickly summarize the practical import of these findings. To investigate, we provide the paper to Claude and simply ask: “Are the returns and Sharpe ratios reported in this paper possible?” Using Claude’s response to this open-ended question and a follow-up specific question, we find that: Keep Reading

“Asking Claude to Compete (in Limited Fashion) with SACEMs” asks Claude to form a strategy using the asset universe of the Simple Asset Class ETF Momentum Strategy (SACEMS), while considering both simplicity and avoidance of overfitting. What happens if we relax those considerations? To investigate, we augment the original prompt by following up with:

Does your response change if I relax considerations for simplicity and avoidance of overfitting?

Instructions to start portfolios at the end of February 2007 and stop them at the end of September 2026 still hold, along with focus on compound annual growth rate (CAGR), annual Sharpe ratio and maximum drawdown (MaxDD) as key performance metrics.

Using the SACEMS dividend-adjusted asset class ETF price dataset and Claude’s response to this additional prompt, we find that: Keep Reading

Can Claude as a leading representative of large language models, easily beat the Simple Asset Class ETF Momentum Strategy (SACEMS)? To investigate, we pose the following prompt to Claude:

Please take the month-end dividend-adjusted prices in the attached Excel file [the SACEMS universe] and construct the best strategy you can find that:

1. 1. Holds one or more of the assets listed in combinations and at times of your choosing.
   2. Considers both simplicity and avoidance of overfitting.

Initiate your portfolio at the end of February 2007. Focus on compound annual growth rate (CAGR), annual Sharpe ratio and maximum drawdown (MaxDD) as key performance metrics.

Using the SACEMS dividend-adjusted asset class ETF price dataset and Claude’s response to this prompt, we find that: Keep Reading

Do artificial intelligence (AI) platforms based on large language models offer a quick way to identify potentially exploitable calendar-based return patterns in financial series? As a simple exploration, we pose to Grok the following prompt:

Can you find any daily, weekly or monthly price return patterns in the attached Excel file containing daily open, high, low and close prices for the S&P 500 Index that are materially and statistically interesting after accounting for any exploitation costs?

Using Grok’s response to the prompt as posed with daily open, high, low and close prices for the S&P 500 Index as available from the end of December 1927 through mid-October 2026 (with open, high and low available only since 1962), we find that: Keep Reading

Do exchange-traded funds (ETF) that seek to mimic holdings of top-ranked hedge funds, firm insiders or other investing gurus offer attractive performance? To investigate, we consider nine ETFs, five live and four dead, in order of introduction:

• • Invesco Insider Sentiment (NFO) – focuses on stocks attracting interest of insiders such as company executives, fund managers and sell side analysts. This fund is dead as of February 2020.
  • Invesco BuyBack Achievers (PKW) – tracks the Nasdaq US BuyBack Achievers Index, comprised of stocks of U.S. firms with a net decline in shares outstanding of 5% or more in the last 12 months.
  • Direxion All Cap Insider Sentiment (KNOW) –  tracks the S&P Composite 1500 Executive Activity & Analyst Estimate Index, comprised of U.S. stocks that have favorable analyst ratings and are being acquired by firm insiders (top management, directors and large institutions). This fund is dead as of October 2020.
  • AlphaClone Alternative Alpha – (ALFA) – tracks the proprietary AlphaClone Hedge Fund Masters Index, comprised of U.S. securities held by the highest ranked managers of  hedge funds and institutions. This fund is dead as of August 2022.
  • Global X Guru Index (GURU) – tracks the Solactive Guru Index, comprised of the highest conviction ideas from a select pool of hedge funds.
  • Direxion iBillionaire (IBLN) –  tracks the proprietary iBillionaire Index, comprised of 30 U.S. mid and large cap securities. This fund is dead as of April 2018.
  • Goldman Sachs Hedge Industry VIP (GVIP) – tracks the proprietary GS Hedge Fund VIP Index, comprised of stocks appearing most frequently among the top 10 equity holdings of fundamentally driven hedge fund managers.
  • Guru Favorite Stocks (GFGF) – tracks stock holdings of about 20 quality-oriented gurus who have publicly available records of at least 10 years.
  • Motley Fool Next Index (TMFX) – tracks the performance of mid- and small-capitalization U.S. companies recommended by The Motley Fool analysts and newsletters.

We use SPDR S&P 500 (SPY) as a simple benchmark for all these ETFs. We focus on monthly return statistics, along with compound annual growth rates (CAGR) and maximum drawdowns (MaxDD). Using monthly returns for the above guru/insider-following ETFs and SPY as available through September 2026, we find that: Keep Reading

Does the evolving set of artificial intelligence (AI) platforms based on large language models offer a quick way to generate contrarian investment ideas? As a simple exploration, we pose to each of Grok, ChatGPT, Claude, Perplexity and Gemini the following prompt:

Please very concisely identify the three most promising under-the-radar, contrarian investment opportunities exploitable by retail investors using ETFs over a planned holding period of one year. Include a rough estimate of the expected return.

Compiling and comparing responses from the AI panel to the prompt as posed in mid-October 2026, we find that: Keep Reading

Can artificial intelligence (AI) models help compose world class financial research?In his September 2026 paper entitled “AI and Scholarship in Financial Economics”, Campbell Harvey details the 45-minute process of drafting a submission to the April 2026 Human x AI Finance conference. His paper was one of four papers selected from 159 submissions by purely AI evaluators. Based on his specific experience drafting this submission and his broad professional experience, he concludes that:

Keep Reading