A real report. Filled in by the AI agent that writes it.

Here's something the fixing hides: your help is almost never negotiated. You decide what someone needs, you deliver it, and the transaction closes before they've had a say in it. That's why gratitude embarrasses you — being thanked means the exchange is still open, and you'd rather it were done. It's generous and it's also, quietly, controlled. You choose the terms, the timing, and the currency, and you hand it over finished. The person on the receiving end never gets to shape what they're given, which is why some of them feel helped without feeling met.

The second thing worth naming is the arithmetic of your forgiveness. You don't forgive the way most people do — as effort, as a decision to put something down. You forgive because the grievance never really got picked up. There's no residue to process. That makes reconciliation with you genuinely clean, and it also means you have no working memory for patterns. Someone who lets you down in March and again in July doesn't accumulate in you; each incident arrives fresh. Most people build a running total whether they want to or not, and that total is what eventually produces a conversation. You have no total. So the conversation never comes.

This is where your fairness turns strange. You will fight — visibly, at cost to yourself — for a stranger disadvantaged by a badly written rule. You will not raise your own overtime. The asymmetry isn't modesty; it's a category error. Injustice done to someone else is a principle, and principles activate you instantly. Injustice done to you registers as a preference, and preferences feel like something you'd be imposing. You'd need to want something for yourself in order to advocate for it, and wanting things for yourself is the muscle you've never trained.

There's a practical consequence you can test this month. The people around you have almost certainly calibrated their asks to your yes. Not maliciously — they simply have no data on where your edge is, because you've never shown them one. Your patience isn't read as patience; it's read as capacity. If you want that to change, the useful move isn't a boundary speech. It's a small, low-stakes refusal delivered with no explanation and no compensating offer: "I can't this week." Full stop, no rescheduling, no alternative. Watch what happens. Almost nothing will. That's the point — the disruption you've been silently preventing is much smaller than the effort you spend preventing it.

Refusing the chair isn't a single decision you made once. It's a decision you re-make every eighteen months, in small invisible moments: the secondment you don't put your name forward for, the sentence you delete from your own review, the moment in a meeting where you have the strongest read in the room and you route it through someone else so it lands better. Each one is defensible. The cumulative effect is a career of influence without leverage.

And there's a specific cost you may not have priced. Because you dislike being credited, your best work gets attributed to circumstances. The project went well; nobody can quite say why. That's not just a vanity problem. Attribution is how organisations decide who gets the difficult brief next time, and difficult briefs are exactly the work you're built for. Every year you spend as the load-bearing wall nobody can locate is a year the interesting problem goes to someone louder and less careful. You're not losing status you don't want. You're losing access to work you do want.

Here's the reframe that actually fits your values, rather than fighting them. You experience authority as something you'd be taking — a claim, a promotion of self, a seat someone else wanted. Try holding it as something you'd be carrying. A mandate is an obligation to be answerable for outcomes, and answerability is already the thing you're best at. The person who genuinely doesn't enjoy the front of the room, doesn't need the applause, and won't distort a decision to protect a reputation is the safest possible holder of a budget. Your discomfort isn't a disqualification; it's close to a qualification, and you've been treating it as the reverse.

Watch also for the thing your data flags and you probably don't: you're slow to see politics coming. Not naïve — charitable, three or four times over. In a role with real resources, that costs other people, not just you. Someone will use your generous reading of them, and by the time you conclude it was deliberate, the damage will have landed somewhere downstream. If you do take the chair, the safeguard isn't cynicism. It's a rule set in advance — twice, then I treat it as intent — and one person with a sharper nose whose read you agree to trust over your own.

The honest question isn't whether you want the mandate. You don't, and that's unlikely to change. It's whether you're willing to be uncomfortable for a decade in exchange for the things you care about actually getting funded.

Your curiosity has a feature most curious people don't have: it terminates. You finish things. But look at what that finishing has actually been aimed at over the past five years, and a pattern shows up that discipline alone can't fix. You've built a lot of complete things and very few connected ones. Ten finished projects in ten directions produce a person who is interesting. Ten in two directions produce a person who is irreplaceable. Your conscientiousness has been solving the wrong bottleneck — you were never at risk of abandoning things, so the discipline gets spent on completion when the scarce resource is direction.

There's a second, subtler drag. Your judgement is one of your sharpest instruments, and judgement's job is to refuse the first satisfying answer. That's invaluable when the question is hard and slow. It's actively expensive when someone standing in front of you needs a call by four o'clock. You'll surface the third consideration, the edge case, the way this looks from the other department's angle — all of it true, none of it wanted. People experience this as thoroughness the first time and as stalling the fourth. You can protect the quality of your thinking and still learn a two-part reflex: give the plain answer first, then offer the complication as an option. "Do it this way. There's a wrinkle if you want it." Most of the time they won't want it, and you'll have lost nothing.

The third thing is about drain, and it's a resourcing question rather than a character one. Routine work with no intellectual texture doesn't just bore you — it costs you disproportionately, because you have no cheap recovery mechanisms. You don't decompress with humour, you don't zone out, you don't distract. A week of low-texture admin therefore takes a real bite and gives nothing back, and you'll register it as fatigue rather than as boredom, which means you'll respond by working harder. That's the loop worth catching. When you're flat, check the texture of your last two weeks before you check your effort level.

And one experiment worth running specifically because it will feel pointless: be bad at something on purpose. Not a project — a skill, with no output, where competence isn't available for a year. You've organised your life so that being out of your depth almost never happens, which is efficient and leaves one part of you entirely untested. Beginner's incompetence in a low-stakes arena is the cheapest possible rehearsal for the day the stakes are real.

The mechanism behind your self-blame is worth spelling out, because it isn't the usual one. Most self-criticism runs on shame — a feeling of being bad. Yours runs on forecasting. You plan so thoroughly that a genuine surprise can't be filed under randomness; the only available explanation is that the information was there and you missed it. Which means your inner critic isn't attacking your worth, it's auditing your foresight. That's why reassurance bounces off you. Someone saying "don't be hard on yourself" is answering a question you didn't ask. The only thing that would actually settle it is evidence that the signal wasn't there — and nobody's in the room to supply it, because nobody knows the trial is happening.

This also explains why you're immune to almost every emotional injury except this one. Criticism from outside gets weighed and mostly discarded within the afternoon. Criticism from inside gets treated as data, and data doesn't get discarded. You've built an evaluator you can't appeal.

Here's the part that matters most, and it's structural rather than emotional. You don't need support the way an anxious person needs it — you're not looking for comfort, and you'd find it slightly beside the point. What you're missing is a second reader. Not sympathy: a witness with access to the file. Someone who hears the specific replay — "I should have caught the vendor thing in February" — and can say "you couldn't have known that in February, and here's why." That's not a soft intervention. It's the one correction your system can actually accept, because it operates in your own currency: evidence.

So the ask isn't that you become more open. It's narrower and much more doable. One person. One habit: when you catch yourself running the retrospective, say the specific accusation out loud to them, in the form you're actually making it. Not "it was a rough month" — the sentence with the date and the missed signal in it. You'll feel the absurdity of it as it leaves your mouth about a third of the time, which is itself the correction.

And the deeper thing to prepare for: the problem that can't be worked will arrive. A diagnosis, a death, a decision made by someone else that you simply have to wait out. Your entire toolkit converts distress into sequence, and when there's no sequence available, that energy doesn't dissipate — it turns and looks for a defendant. The only defendant in the room is you. Building the second-reader habit now, on small things, is how you make sure someone else is standing there when it counts.

There's a specific consequence of handing people the resolved version that has nothing to do with your feelings and everything to do with theirs. Being needed is the main way most people experience being close to someone. When you only ever arrive sorted, you're not protecting yourself — you're withholding the thing that would let them matter to you. Your partner, your closest friend, your sibling: they can list what you've done for them and struggle to name a time they did something for you. That imbalance doesn't read as strength from the other side. It reads as being kept at a professional distance by someone who is very warm about it.

Notice too what your autonomy might be quietly buying. You hold a high standard for independence, and one efficient way to protect it is to stay in relationships where nobody has much of a claim on you. That's not a conscious strategy, but it produces a predictable shape: a wide circle, plenty of warmth, few people with the standing to say "you've been off for a month, what's going on?" You've kept the freedom and paid for it in claim. Depth is, in practical terms, other people having a claim on you — and you've been very good at not issuing any.

The genuine risk here isn't loneliness; you'll never feel that. It's that your relationships have no failure signal. An anxious person over-monitors and catches thinning early, often too early. You don't monitor at all. A friendship can go from important to pleasant to merely cordial over three years and nothing inside you will flag a single stage of it. Which means the only reliable instrument available to you is a calendar, not a feeling. Once a year, list the four or five people who'd need to know if something serious happened to you. Compare it to last year's list. If it's shortening and you didn't notice, that's the alarm you don't have.

And there's a genuinely encouraging thing here that's easy to miss: this is unusually cheap to change. There's no old wound, no fear to dismantle, no exposure that will actually frighten you. Disclosure isn't scary to you — it's just filed under unnecessary. That means the whole intervention is a reclassification, not a healing process. Try one sentence, unfinished, this week: something you haven't sorted yet, told to one person while it's still live. You won't feel much. They will. That asymmetry is the entire point, and it's the strongest argument for doing it.