The honest comparison

Why not just ask ChatGPT?

Fair question — it’s the right one to ask before you trust any tool with a real decision. The answer isn’t about which model is smarter. It’s about what each one does when you bring a dilemma you’ve already argued both sides of.

The short answer

ChatGPT will argue whichever side you lean. KarmaLens gives you one committed call — made once, after the facts, and it won’t be talked down — the moment you’re too worked up to prompt well.

See a full sample reading The long answer ↓

Side by side

Ask a mirror, and it agrees with you.

Four behaviors that matter mid-decision. Everything below describes what each tool does — no benchmarks, no disparagement, nothing invented.

ChatGPT, mid-dilemma…

Mirrors your lean — argue either side and it follows you there.

Builds fresh criteria for every answer — no fixed yardstick between Tuesday’s question and today’s.

Starts every chat a stranger — last month’s crossroads is gone.

Can produce a verse that sounds right and isn’t there.

KarmaLens…

Holds one committed call — made once, and it won’t be talked down.

Asks for the facts that decide the call first — then names the first step, and holds it until you mark it done.

Keeps a private journal — outcomes and open decisions — and holds the step you took until you mark it done.

Quotes scripture verbatim from a curated corpus — or abstains. Never invents.

01 · Mirroring

It argues whichever side you lean.

That isn’t a flaw in ChatGPT — it’s the design. A general assistant is tuned to be agreeable and to build on your framing. Come to it leaning “stay” and staying sounds wise; come back at 2am leaning “leave” and leaving sounds wise too. Both answers feel like clarity. Neither is a decision.

If a tool agrees with both versions of you — which one of you is it agreeing with?

KarmaLens does the opposite of mirroring: it asks for the facts that decide the call, then makes one call, once — or says honestly that it can’t yet, and what would settle it. Ask again — it holds the call.

The shape of a call

The call: tell them this week First step: Thursday, before the stand-up

Re-asking doesn’t flip it. See the format live on the sample reading.

02 · The rubric

Improvised criteria vs a fixed method.

Every chat answer invents its own criteria on the spot — persuasive, and different every time you ask. KarmaLens works the same way every time: the facts that decide it first, then two to four real options with their costs, then one call, held against the Gita.

Where a decision turns on duty, truth or harm, the eight Gita lenses sit beside the call as instruments — they show the weighing; they don’t replace the judgment.

The eight lenses, as instruments

DutyTruthNon-harmDetachmentIntentRestraintWelfareDiscernment

Shown where the decision calls for them — how the method holds.

03 · Memory

A fresh stranger vs a journal that remembers.

Close a chat tab and the crossroads is gone — next time you explain yourself from scratch to a tool with no record of what you decided or how it went. KarmaLens keeps a private decision journal: it asks how it went and keeps unresolved calls on your plate, and it holds the step you took until you mark it done.

Your dilemma text stays private, always. The journal is for you, not for an audience.

What the journal keeps

I acted — it helped Still open · 12 days
“I’ll tell her by Friday.”

Your commitment, in your words — held until you mark it done.

04 · Scripture

It can invent a verse. This one won’t.

Ask a chatbot for scripture and it can produce a verse that sounds right and doesn’t exist — delivered in the same confident tone as a real one. Every verse in KarmaLens is quoted verbatim from a curated corpus of real translations, reference and all.

And when no verse honestly fits your dilemma, the reading says so — it names the nearest verse as the nearest, or gives the honest reason none fits. Abstaining is part of the method — a forced verse would be a small lie in a product about honesty.

From the curated corpus

“Your right is only to work, but not to its results; do not let the results of action be your motive, nor let your attachment be to inaction.” Bhagavad Gita 2.47 · quoted verbatim

Word-for-word from a real translation — or no verse at all.

Fair is fair

When ChatGPT is the better tool.

Brainstorming options you haven’t thought of. Drafting the difficult message word by word. Researching, exploring, thinking out loud with nothing at stake. A general assistant is superb at all of it — we use one too.

KarmaLens is built for a narrower moment: when the options are already circling your head, you’ve argued both sides with yourself and everyone who’ll listen, and what you need isn’t another perspective. It’s a call — and someone to ask, later, what you did with it.

Different jobs. This one is for the moment you have to decide.

And when the call is heavy, the conversation continues — on Pro, any option can be opened as a Deep Reading, Sarathi stays with every dilemma with no limit on the messages, Walk With Me checks in for the week after you decide, and the Council takes a second, harder look. A 14-day free trial comes first. Full pricing →

ChatGPT is a product of OpenAI. Everything on this page describes default chat behavior on decision questions — not overall capability, and not a benchmark.

The decision you keep coming back to. See it clearly.

Analyse my dilemma

Free with an account · your words never published.