The Boardroom is invest-like's multi-investor stock-debate format. Four named-investor personas plus a skeptic argue any ticker you choose, citing documented framework writings, with the debate itself as the output rather than a single verdict.
Last reviewed:
A traditional AI stock tool gives you one verdict. The Boardroom does the opposite. Pick a ticker; the system instantiates five voices, each grounded in the documented writings of one named investor (or, for the skeptic, the company's own filings). The voices argue the stock under their framework's actual criteria, cite specific passages they are reasoning from, and disagree publicly. The output is the transcript of that argument, not a single buy or sell tag. The structured disagreement is what makes the output more useful than a single LLM-generated paragraph.
Each named-investor voice is anchored in a corpus of their published writings. The skeptic is anchored in the company's own filings.
The quality compounder
Argues from the moat-and-durability angle. Cites Berkshire shareholder letters spanning 1965 to 2024, and from his early partnership letters where the framework first appeared. Will dismiss a fast grower with no moat; will defend a low-growth incumbent with a fortress balance sheet and a 30-year demand history.
The defensive deep-value voice
Argues from the statistical-cheapness angle. Cites The Intelligent Investor and Security Analysis. Will reject growth premiums of any kind, including ones Buffett tolerates. Asks whether the price gives a clear margin of safety against tangible book and earning power, ignoring whatever narrative the other voices are telling.
The growth-at-reasonable-price voice
Argues PEG and earnings velocity. Cites One Up on Wall Street and Beating the Street. Will champion fast growers Graham and Munger would pass on, and will dismiss compounders trading at a multiple ahead of their growth rate. Looks at insider buying and institutional ownership as live signals.
The Magic Formula voice
Argues purely from the ROIC and earnings-yield ranking. Cites The Little Book That Beats the Market and his Gotham Capital writings. Does not care about moat, management quality, or narrative; cares only whether the ticker would land in the top decile of his ranked screen across the US universe.
The dissenting fifth voice
Not a named investor. The skeptic looks for what the other four are not saying: accounting red flags, customer concentration, regulatory tail risk, management actions that contradict the IR deck. Cites the 10-K, proxy statements, and any contradiction visible across the company's own filings. Disagreement is the skeptic's only job.
The personas are not free-form character generation. Each named voice draws on a documented corpus that has been chunked, embedded, and indexed. For Buffett that is the Berkshire Hathaway shareholder letters (1965 to 2024) and the early Buffett Partnership letters. For Graham it is The Intelligent Investor and Security Analysis. For Lynch it is the two trade books. For Greenblatt it is The Little Book That Beats the Market and the Gotham public commentary. The skeptic draws on the company's 10-K, 10-Q, and proxy statements.
When a voice makes a claim, it retrieves the relevant passage from its corpus first, then reasons forward. The output includes the cited passage inline so you can read what the named investor actually wrote on the point. This makes the output auditable in a way a pure impersonation prompt is not: you can check whether the quote is real and whether the reasoning derived from it is faithful.
A single AI verdict on a stock can be wrong in a way the reader can never detect. A debate cannot. If Buffett and Lynch reach the same conclusion via different reasoning paths, the agreement is meaningful. If they reach opposite conclusions, the disagreement itself shows you the live argument: Buffett would call the company a fragile growth bet; Lynch would call it a compounder priced below its earnings velocity. Either way you exit with a clearer mental map than a single paragraph can give you.
The format is also more honest about model uncertainty. An LLM that has to defend a position in front of four other voices answerable to specific named-investor criteria is forced to produce a more careful, more defensible argument than one writing a free-form summary. The other voices act as adversarial reviewers within the same generation pass.
Every Monday: the stocks that newly pass 5 or more of the 7 legendary-investor frameworks (Buffett, Graham, Lynch, Greenblatt, Munger, Fisher, Smith), plus every grade change that matters. The 5-year track record of this screen is published openly. One email a week, unsubscribe anytime.
Pick any US-listed stock; the five voices argue it on the fly with full citations and a downloadable transcript.
Educational only. invest-like is not a registered investment adviser; the named-investor voices are retrieval-grounded simulations and do not represent the actual views of any individual. Always do your own research and consider your individual circumstances.