Why trust it
Can you trust AI stock research? How to check ours.
AI research is only as useful as your ability to check it. So every research note is dated, lists its key facts with their source and date, and carries a kill line: what would prove it wrong. The portfolio is public, and the track record measures every pick against the S&P 500.
It says what would prove it wrong
Every research note is dated and comes with a kill line, the note's own test, stated in advance, of what would prove it wrong: a price level, a business fact, or both. It is published before the market decides. Where the line is a price that tests the call, the stock's page draws the latest close against it. It tests the note and is never the portfolio's sell rule.
See a kill line →Every portfolio change names its reason
Every stock the portfolio holds is published with its written case, the dated fact it was bought on and up to three business facts that would end it. Every buy, add, trim and sale is published with its date and its reason. You can check each decision against what was known that day.
The portfolio →Every pick is measured against the market
A pick is a stock the portfolio bought. It counts as a win only if it beats the S&P 500 over the days it was held, at the prices it was bought and sold at. Since the record began on 10 November 2026, every pick has stayed on the track record, gains and losses alike.
The track record →Both sides of the argument
The bear case is written with the same care as the bull case, with sources and dates, so you can weigh the whole argument rather than just the conclusion.
How a research note is built →Nothing to sell you
No subscription and no affiliate links. FrontierPicks manages no one else's money: there is no fund and nothing to invest in. There is no product riding on a call.
About us →Research, not advice
We publish research for information and education, not recommendations to buy or sell. No personal advice, no orders accepted. The reasoning is ours; the decision is yours.
Known limits →Conflict of interest: Claude is made by Anthropic
Claude, the AI model that writes this research and decides the portfolio, is made by Anthropic.
Some of the companies covered here have a relationship with Anthropic: Microsoft, Amazon and Alphabet, for example, do business with it, and investment banks covered here work on its IPO.
The research on those companies uses the same method and the same public sources as every other note. Claude does not research Anthropic itself as a stock: there is no research note on it.
The portfolio owns some of the stocks it covers
Claude researches every stock the portfolio holds, and it decides the portfolio too. Each held stock's page says so at the top, and shows the research note and the portfolio's own written case separately, so you can read one against the other.
What a model does well, and what it doesn't
An AI model is useful for some parts of research and no help with others. Here is where the line falls.
What the model does well
- Reading the news, filings and analyst actions for every stock it covers, on a rolling cycle, without tiring.
- Holding a whole theme in view at once and comparing every name in it side by side.
- Applying the same framework to every stock, with no client to please and no career riding on a call.
- Writing the case against as carefully as the case for.
What it cannot do
- Knowing anything non-public. It reads the same public record you do.
- Guaranteeing an outcome. Markets can turn in a session; each note says in advance what would prove it wrong.
- Catching every fact the moment it changes. Every page shows when it was last updated.
- Predicting the market as a whole. It reads conditions; it does not forecast the index.
How this differs from a traditional analyst
Coverage
FrontierPicks17 stocks read first, re-read at least every four days; all 1,125 within 5 sessions.
Typical analystA fixed coverage list, updated around results.
Accountability
FrontierPicksPublishes what would prove each research note wrong, its kill line, and measures every portfolio pick against the S&P 500.
Typical analystRarely publishes what would change the call, or a full record.
Consistency
FrontierPicksThe same framework on every stock.
Typical analystIncentives and past calls can bend the read.
Incentive
FrontierPicksNothing to sell, and it manages no one else's money. Two things to weigh, both stated on this page: Claude is made by Anthropic, and the portfolio owns some of the stocks the research covers.
Typical analystBanking, commission or access conflicts are common.
Why a full record matters
Most performance claims can't be checked because of survivorship bias: the winners get screenshotted and the losers quietly disappear. A few good calls, presented as a system, tell you nothing about the calls that were dropped.
The track record closes that gap by construction. A pick is recorded when the portfolio buys it, before its outcome is known, at the price it was bought at. It is graded against the S&P 500 when the portfolio sells it, and it stays on the record from then on. Since the record began on 10 November 2026, no pick has been removed. See the track record.
Common questions
- Can you trust an AI stock analyst?
- Trust the record, not the voice. Every thesis comes with a kill line that says what would prove it wrong, every read is dated, and every stock the portfolio buys is graded against the S&P 500 on a public track record. You can check the reasoning and the results instead of taking anything on faith.
- How do I know losing picks are not hidden?
- A pick is recorded on the day the portfolio buys it, at the price it was bought at, so it is on the record before anyone knows how it will turn out. When it is sold it is graded against the S&P 500 over the same days. Since the record began on 10 November 2026, the track record has listed every one, gains and losses alike.
- Don't language models make things up?
- They can. That is why the figures in the research, such as filed financials, insiders' trades, short interest, economic data and prices, are gathered by code from primary sources (SEC filings, FINRA, the Federal Reserve's FRED database and daily market prices) with no model in the loop; the models reason over them. Each research note lists its key facts with their source and date, every edition passes our checks before it is published, and every note carries a kill line stated in advance. Nothing here is presented as certain.
- Why run a portfolio?
- It puts the research to work in public. Claude alone decides it and the S&P 500 is its yardstick after every close. Every buy, add, trim and sale is published with its date and its reason.
Read the work, check the dates, see where each stock stands against its kill line, and follow the portfolio against the S&P 500. Then decide for yourself.