AlphaMaven AlphaMaven

Rob Carver

Reported by Top Traders Unplugged
Nov 21, 2026· 2 min read
Save Post LinkedIn Email
Key Takeaways
  • • Top Traders Unplugged Episode 427 (dated 21 November 2026) hosted by Niels Kaastrup-Larsen and Rob Carver examines whether AI can be trusted to evaluate its own trading strategies.
  • • The discussion centres on risks of AI-driven research, including overfitting, crowded trades, and backtests that look convincing but lack validity.
  • • The episode covers trend following, portfolio protection, CTA performance, and whether AI-generated ideas can be genuinely original.
  • • Self-driving portfolios and look-ahead bias in backtesting are identified as areas where human oversight remains essential.
Summary
In Episode 427 of the Top Traders Unplugged podcast, dated 21 November 2026, hosts Niels Kaastrup-Larsen and Rob Carver address the promise and pitfalls of AI-powered investing. The episode opens with the central question of whether artificial intelligence can be trusted to assess its own trading strategies, noting that a model can build a strategy, write its backtest, and declare the results brilliant without independent verification.

The hosts examine several risks associated with faster AI-driven research. These include overfitting, where strategy optimisation drifts into fitting historical noise, and the possibility that widespread use of AI-powered research could lead to market crowding and declining alpha. The discussion also considers whether AI can generate genuinely original investment ideas, as opposed to recombining existing approaches.

The conversation extends to trend following and portfolio construction. Segments address recent hedge fund developments and the latest CTA performance, how Rob Carver's diversified trading portfolio is navigating volatile markets, and how trend following might complement a concentrated investment portfolio. Market-specific momentum models and the trade-off between complexity and robustness are also discussed.

Later segments focus on methodology and governance. The hosts contrast traditional approaches to designing and testing trading strategies with the dangers of self-driving portfolios and AI-generated backtests, and outline where AI can add value and where human judgement remains essential. The episode closes with practical guidance on building more trustworthy backtests free of look-ahead bias.

The source material is a podcast episode page that includes promotional content for DUNN Capital research and a free investment book guide; no specific fund flows, deal values, or quantitative performance figures are reported in the source.

Go deeper than the headline

Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.

Start your free 7-day trial

Sources

1 outlet

AlphaMaven summarised this story from the reporting below. Each link opens the original article.

Top Traders Unplugged Rob Carver