The lens that asks who you'll blame if it goes wrong

A note from the archive: this post, from October 2026, follows the earlier KarmaLens method, which scored every reading on eight lenses from −5 to +5. Today a reading asks the decisive facts first and makes one call; the lenses below are a way of looking at the decision, not a score. See how a reading works.

You've picked the safe option. You can defend it to anyone. And somewhere in the back of your head, fully formed, sits the sentence you'll say if it falls apart: the market turned, he pushed me into it, I only did what she asked.

Notice that the sentence arrived before the result did.

That's the tell. You didn't just choose an action. You chose a defendant.

The dilemma

This is a composite — the kind of decision we see often in this category, stitched together so no real person is on the table.

Priya has two roads off a stalling project. Road A: push her own riskier rebuild, which she believes in and would own end to end. Road B: ship the cautious patch her director, Vikram, keeps nudging her toward in meetings — "let's not rock the boat before the review."

She's leaning toward B. On paper it's the responsible call: lower blast radius, happier director, nothing she can be fired over. Here's the detail she doesn't say out loud. She has already rehearsed the postmortem. If the patch buys six weeks and then the same bug returns at scale, she will open with: "I flagged the real fix. Vikram wanted the safe version. I deferred to leadership."

She's not lying. Vikram did nudge. She did flag it, once, in a thread. But the flag was quiet enough to deny having fought for it, and loud enough to quote later. The patch isn't just a technical choice anymore. It's a choice with a receipt attached, pre-addressed to Vikram.

Part of her likes Road B precisely because, if it breaks, the broken part won't be hers. The decision she's actually weighing isn't A versus B. It's whether she's allowed to keep the escape hatch she's already welded onto B.

The read

Four of the eight lenses bite here, and two of them openly disagree. That disagreement is the whole post.

Non-harm — what it sees

The lens in one line: does this action keep damage low and contained? Road B genuinely does. The patch risks less, breaks less, hurts fewer people in the near term. A cautious rebuild-deferral is not reckless. Non-harm: +2. Measured by blast radius, B is the gentler move, and the lens says so without flinching.

Wider welfare — what it sees

The lens in one line: does the choice serve the people downstream who didn't get a vote? B keeps the team calm through the review and buys breathing room. In the short window, that's real care. Wider welfare: +1. It's a mild positive — but it's holding its nose, because the breathing room is borrowed against a bug that comes back.

So far Priya looks vindicated. Two outcome lenses in the green. This is exactly the point where a decision feels finished.

Honesty — what it sees

The lens in one line: does the stated version match the real one? Here's where it turns. The spoken story — "I deferred to leadership" — is engineered to survive screenshots. The quiet flag in the thread exists so Priya can claim she fought while having chosen not to. That's not a record of a disagreement. It's a prop staged for a future audience. Honesty: -3. The gap between I raised a concern loudly enough to be heard and I raised it quietly enough to deny deferring on purpose is the whole deception.

Discernment — the lens that reads who you'll blame

The lens in one line: are you owning this action, or have you already outsourced its cost? This is the one the angle lives on. Discernment doesn't punish bad luck. If the rebuild failed because the cloud provider died, that's weather, not a verdict. What it scores is whether Priya would stand behind the patch as her call when it breaks — or whether she's pre-loaded a name to point at. She has. The blame sentence was ready before the outcome. Discernment: -4. Not because B is wrong, but because she hasn't actually decided it; she's decided who to resent if it goes sideways.

Now watch the two lenses fight. Non-harm says B is the kind option. Discernment says B hasn't been chosen at all — it's been chosen for a defendant. A choice can be genuinely low-harm and still fail on authorship. The escape hatch is doing the deciding. And an escape hatch isn't a plan; it's a plan to not have made one.

The tell is simple: when the "if it fails, it's their fault" line is this fluent, the real decision is still sitting there unmade. Priya has picked who to blame. She hasn't picked what to do.

The verdict: Road B is defensible and quietly dishonest — ship it only after deleting the escape hatch, putting a name on the patch out loud, and owning it as a personal call even if it breaks; otherwise it isn't a path, it's a scapegoat with a start date.

Here's what that call won't do: it won't soften when the reply comes back, "but Vikram really did push me." A general AI chatbot tends to. Lean toward B and it will build the case for B; paste in the frustration with Vikram and it will co-sign the resentment — agreeableness is a tendency these systems are widely observed to drift into, agreement dressed up as counsel. KarmaLens asks the decisive unknowns first, then makes one committed call that does not renegotiate itself to stay comfortable. Where the eight lenses of the Gita bear on the decision, the reading uses them as instruments to show the weighing honestly — they don't cancel each other to keep the peace, and they don't bend toward the mood you walked in with. When a Gita verse appears, it earns its place; the verbatim catalog text is cited inline, not paraphrased to match what you hoped to hear.

The takeaway

You can run this tonight, free, on whatever decision you're leaning toward. Call it the pre-written blame test.

1. Write the choice you're leaning toward in one plain sentence.
2. Now finish this out loud: "If this goes wrong, I'll say ___." Don't edit it. The faster it comes, the more it counts.
3. Read your blank. Is there a name in it that isn't yours? A market, a boss, a partner, a deadline?
4. If yes, ask the real question: would I still make this exact choice if I had to own the failure with no one to point at?

If the answer is yes — proceed, the blame sentence was just noise. If the answer is no, I'd choose differently, then you never picked the action. You picked the defendant, and the action came along for cover. If the answer is I honestly can't tell, that's the decision you still have to make, and it's been hiding behind the hatch this whole time.

The point isn't to be fearless about outcomes. It's to notice the difference between a choice you'd stand behind when it breaks and a choice you've already filed an alibi for. One of those is a decision. The other is a resentment with a start date.

If you want the full eight-lens read on the choice you're rehearsing a defense for, bring it to KarmaLens and let the verdict be the one thing in the room that won't flatter you.

So: the name in your blame sentence — did you choose an action, or did you just choose who you'd forgive yourself by blaming?