What is Precis?

Longevity advice contradicts itself, and no one can keep up with all of it. Precis is an independent research analyst. It reads researcher interviews, gives each claim one verdict worked out from every mention, and shows the receipt for every line: the video and the minute.

No affiliates. Nothing to sell you. Every claim points to the video and the minute.

The facts

What it is

An independent research analyst for longevity. Each claim gets one verdict, worked out from every counted mention, and every mention behind it is on the page.

Who runs it

Arun Agrahri, a technologist. One person, no team, no medical license.

What it reads

Built from 3,700+ hours of recorded interviews with 380 researchers and doctors, plus 48,500+ research studies. Updated Oct 14, 2026. Someone counts as a researcher or doctor only when an episode title puts Dr., Professor, PhD or MD next to their name. Video and podcast transcripts only.

What it publishes

44 consensus reports across 50 tracked supplements, plus dosage, comparison and goal pages.

How a claim is checked

Every published dose and form has to match a line in a real transcript. A verification script re-reads the corpus and flags any number, form or product name nobody said. We do not publish until it comes back clean. If it is not on tape, it does not ship.

What a citation looks like

The person who said it, the video, and the second they said it, so you can open the source and hear it yourself.

What it costs

Every report's public page is free with no account. A free account opens the full reports, with doses and timing. Pro is $9/month or $79/year and pays for the AI: unlimited Ask, chat, and photo and lab reads.

How it makes money

Only readers pay us: Pro subscriptions, and a one-off $49 health assessment report. No affiliate links, no brand sponsorship, no storefront. Nobody pays to score well. How we stay independent.

Who this is not for

Anyone who wants a diagnosis or a personal protocol: Precis reports what was said in public, not what you should do. Anyone who wants a review of the published research: we read what researchers say on camera and the studies they cite there, which is a narrower window on purpose. Anyone who wants one clean answer: on a contested topic the verdict says Mixed or Conflicting and shows each side. And anyone shopping for a brand to buy, because we do not name bottles.

We're allowed to ask for the evidence.

We read many researchers and work for none. Our purpose is bigger than any one of them. No brand, sponsor or show pays us. The people who read Precis do.

Who runs this

Precis is built by Arun Agrahri, a technologist applying data and technology to longevity questions. I am not a doctor, a dietician or a licensed healthcare provider. I take longevity seriously and I believe the answers are in the data, not the hype. Find me at agrahri.com or @arunagrahri.

How reports are created

Every report follows the same five steps:

Transcript collection — we pull and process the transcripts. Built from 3,700+ hours of recorded interviews with 380 researchers and doctors, plus 48,500+ research studies. Updated Oct 14, 2026.
Summarization with stance extraction — each video is summarized into structured claims tagged by stance (recommends, cautious, disagrees), and every claim carries the video ID and the second it starts.
Verification against the tape — a verification script re-reads the transcripts and flags any published dose, form or product name that is not in one of them. Quoted passages are re-checked against the timestamp they claim. Numbers that survive are numbers somebody said.
The verdict — each claim gets one verdict, worked out from every counted mention. The verdict is Strong consensus, Leaning, Mixed, Conflicting or Not enough evidence, and nobody types it in.
Report assembly — stances, dosage protocols, citations and the points of disagreement are compiled into the published report.

The rule underneath all of it: if nobody said it on tape, it does not get printed. That is why we cannot answer some questions — and why we say so on the page instead of filling the gap.

How channels are chosen

Channels are picked on three criteria: reach (millions of subscribers), depth (they name specific doses, mechanisms and protocols) and credibility (academic or clinical background).

Not every guest is a researcher, and the list includes journalists and authors. We count someone among the researchers and doctors only when an episode title puts Dr., Professor, PhD or MD next to their name.

Data at a glance

3,700+

Hours of recorded interviews

380

Researchers and doctors

48,500+

Research studies

Reports scored

Built from 3,700+ hours of recorded interviews with 380 researchers and doctors, plus 48,500+ research studies. Updated Oct 14, 2026. Counts are read from the corpus and rounded down, so the "+" is always true.

What Precis is not

Precis is an information tool, not a medical provider. We do not diagnose, treat or prescribe. Everything here is analysis of publicly available video content, and none of it substitutes for professional medical advice.

Always consult your healthcare provider before starting, stopping or changing any supplement or health protocol.