AI equity research · explained
Stock research a model can be held to.
AI equity research is stock analysis written by frontier language models instead of human analysts. Done well, it pairs two things a human desk rarely manages at once: it reads the primary sources behind every stock it covers in depth, and it is falsifiable: each call carries a level or event, set in advance, that would prove it wrong, and a real closing price settles it. Every dossier here is dated and two-sided, across 17 themes. All 225 published dossiers include both a dated thesis and a kill line, the price that would prove it wrong.
What AI equity research actually is
A frontier model reads the news, company releases, filings and analyst actions behind each stock it covers. It clusters names by the narratives moving them, then writes a structured thesis per name — a current read, a bull case, a bear case, and the one thing that would prove it wrong.
That is the easy half. The hard half — the half almost no one ships — is making the output accountable: dated, two-sided, and falsifiable, so it can be graded later instead of quietly forgotten. Research you can't be wrong about isn't research; it's marketing.
Why a public, falsifiable track record matters
Two failure modes define the genre. The first is the black box: a confident call with no stated reason you can check. The second is the disappearing scorecard: picks published, losers never mentioned again, and "performance" quoted as a return marked to a benchmark after the fact.
A falsifiable record closes both. When a thesis ships with the exact level or event that would kill it — written down before the outcome — there is nothing to rationalise later. The call either held or it didn't. Every medium-or-higher conviction pick is then graded against the S&P 500 over its own window, under a rule written down before the first pick opened. Each call is on the record, dated.
Can a language model actually pick stocks?
The research layer is where a frontier model earns its place. Reading a stock's recent news, releases, filings and analyst actions, then compressing that into a dated, two-sided thesis with one explicit kill condition, is a synthesis problem — and a frontier model does it consistently, name after name, without the fatigue or the favourites that bias a desk. It writes the bear case as carefully as the bull case because it has no position to defend.
Reading well is not the same as being right, and the difference is the only one that counts. The test is not whether the thesis sounds sharp; it is whether the call beats the market over the window it names. That is a question the prose can't answer and the record can, which is why every dossier carries a kill line a closing price can settle, and why every pick is graded under one published rule. The scoreboard is where that question gets settled rather than asserted.
What to ask of any AI stock service
A model that writes a sharp thesis has shown it can read; whether its calls beat the market is a question only a graded record can answer, over enough picks to tell skill from luck. That is what the FrontierPicks scoreboard is for: it started on 1 October 2026, and every pick from that day on stays on it.
So the thing to ask of any AI stock researcher is not a return quote but a record that could have gone wrong in public: dated, falsifiable, every call kept whether it worked or not. Most can't show one, because of survivorship bias — the winners are kept and the losers quietly dropped. A record where every call stays visible is the one you can audit.
How to evaluate an AI stock researcher
Six questions separate a record you can trust from one you can't. Apply them to anyone — including FrontierPicks.
Is each call falsifiable?
A specific, published level or event that would prove the thesis wrong — set in advance, not rationalised after.
FrontierPicks:Every dossier carries an explicit invalidation trigger.see it →
Is it scored in public?
Every call graded against a benchmark over its own window under a rule fixed in advance, losers included, not a hand-picked list of winners.
FrontierPicks:Every medium-or-higher conviction pick is graded against the S&P 500 on a public scoreboard.see it →
Is every read dated, and is the grade frozen?
When was this written, and can the author restate the call after the outcome? A research note with no date is unfalsifiable by construction; a gradeable one whose grade can be rewritten is worse.
FrontierPicks:Every dossier and journal entry is dated. A pick's conviction, kill line and thesis are frozen at the close it opens on, and the thesis is fingerprinted.see it →
Is the methodology open?
How names are screened, scored, and sized — stated plainly, not a black box you're asked to trust.
FrontierPicks:The full process is published, stage by stage.see it →
Does it show both sides?
A bull case AND a bear case AND named failure modes — not a one-sided pitch.
FrontierPicks:Every dossier argues both sides and names what would break it.see it →
Is there anything to sell?
A free, falsifiable record aligns the author with being right. A paywall, signal service, or affiliate link aligns them with conversions.
FrontierPicks:No product, no paywall, no affiliate — nothing to sell.see it →
How FrontierPicks does it
FrontierPicks is an AI equity-research analyst built on Anthropic's Claude: Claude Opus writes the research, and Claude Fable decides the model portfolio. We run a five-step process on the New York market calendar: a daily scan of the liquid market by trend, the strongest stocks grouped into themes, a deep read of each stock, a call with the price that would prove it wrong, and grading of every pick against the S&P 500 under a rule fixed in advance. The output is 225 per-ticker dossiers, a regime-tagged daily journal, a macro view rewritten every trading day, a scoreboard of every pick graded against the S&P 500, and a model portfolio published in percent.
It is the answer to the six questions above, by construction: falsifiable triggers, public scoring, dated reads, open methodology, two-sided cases, and nothing to sell.
Common questions
- What is AI equity research?
- AI equity research is stock analysis produced by frontier language models rather than human analysts: the model reads news, company releases, filings, analyst actions and price history and writes a structured thesis per name. The credible version is falsifiable and dated — each call carries an explicit level or event that would prove it wrong — so it can be scored after the fact instead of quietly forgotten.
- Which AI stock research tools publish a track record?
- Most don't — they publish picks and never keep score, or they mark returns to a benchmark after the fact. We publish every thesis with the exact price that would prove it wrong, set in advance, and grade every medium-or-higher conviction pick against the S&P 500 over its own window on a public scoreboard.
- How do you evaluate an AI stock researcher?
- Ask six questions: Is each call falsifiable (a published trigger set in advance)? Is it graded in public against a benchmark, losers included? Is every read dated, and is the call frozen once it is made? Is the methodology open? Does it argue both the bull and bear case and name its failure modes? And is there anything to sell — because a free, falsifiable record aligns the author with being right, while a paywall or signal service aligns them with conversions.
- Can an AI language model actually analyse stocks well?
- For the research layer — turning a stock's recent news, releases, filings and analyst actions into a structured, dated thesis with a bull case, a bear case and a kill line — a frontier model is strong and consistent. But reading well isn't the same as being right. The real test is whether the dated calls beat the market over the windows they name, which is what the public scoreboard measures, pick by pick.
- Is AI equity research investment advice?
- FrontierPicks is not. We publish educational research under the EU and BaFin framework — a transparent, dated record of how a language model reads the market, not a recommendation to buy or sell anything. Always do your own work.
- How is FrontierPicks different from a stock-picking newsletter?
- A newsletter sells conviction and rarely keeps score. We publish every pick with its exact kill line, grade each one against the S&P 500, and run a model portfolio on the research with its returns in public. There is nothing to buy.
- Do AI stock pickers beat buy-and-hold?
- Few have shown it over long periods. A model's edge tends to be strongest over short windows and a handful of names, and over years buy-and-hold is a hard benchmark to beat. Many published claims keep the winners and drop the losers. Answering the question takes a dated record that grades every call against the market — which is what the FrontierPicks scoreboard, started on 1 October 2026, does.