Knowledge
creating
company.

Tacit recovers the judgement behind the decisions people and AI already make, and shows how far you can rely on it. One engine, Tom, reads the decision records a team already keeps and compares the judgement in them with outcomes, and with how the team decides. In underwriting, that keeps an expert's judgement after the person leaves. In the security operations centre, it shows how far to trust the judgement of an AI agent.

Underwriting CyberSOC Diagnostics Nuclear
The Tacit platform

Built to make
the expert judgement
machine-legible.

§ 01 / Thesis

Thesis, briefly.

A read of ~ 2 minutes

AI agents are good at retrieval. They are good at language. In regulated work they fail at the one thing that matters most: judgement.

The published evidence on enterprise AI is now consistent enough to read as a pattern. BCG and McKinsey report 88% adoption against 5% meaningful returns. Gartner reports 85% of projects never reach production. MIT's 2024 study found 95% of enterprise generative AI pilots produced no measurable ROI. The standard explanation is that the models are immature, that retrieval needs to be better — that explanation has been wrong for at least two years.

The agents are failing because the AI industry has been trying to capture expertise as content. Most of the meta decisions that matter have never been written down. The part of expertise that defines the senior expert is not knowledge that can be made larger or more accessible. It is a runtime construct, a live cognitive process that decides, under uncertainty, which heuristic fires. We make that runtime observable, persistent, and queryable for the agents that will inherit the work.

We start in two places, with one engine. Specialty underwriting desks at MGAs and Lloyd's coverholders, where the underwriter sets the price herself. And security operations centres at regulated firms, where AI agents now triage alerts and someone independent has to check their calls.

§ 02 / Diagnosis

The shape of
the problem.

Eight observations

  1. Tacit knowledge does not exist on paper. The documents an LLM can read describe the world after the experts have already left the room.

  2. A thirty-year underwriter cannot tell you, in words, why she declined a submission in four seconds. She just declined it.

  3. A senior radiologist sees the malignancy before they can articulate why. The seeing is the expertise. The words are a post-hoc reconstruction.

  4. A control-room operator detects a drift through a posture, a hum, a feel. The SCADA log shows nothing yet.

  5. These signals do not survive into training data. So they do not survive into the model. So they do not survive into the agent.

  6. What survives is the documentation: the safety bulletins, the discharge summaries, the underwriting guidelines — the thin paper trail of decisions, not the decisions themselves.

  7. LLMs trained on that paper trail are fluent but ungrounded. They are plausible at the centre of the distribution and dangerous at the margins.

  8. The margins are where money, lives and licenses live. Which is exactly where regulated work spends its time.

§ 03 / Engine

What we are building.

one engine · underwriting · CyberSOC

01 — Capture

Five methods

Decision archaeology against historical exhaust, contrastive cohort analysis, perturbation probing, the Structured Interview Agent, and live escalation capture in production. The runtime is opaque to self-report and visible only in behaviour.

02 — Structure

Judgement graph

Every captured node carries a confidence score, a decay rate, and a provenance trail to the named expert and source session. Organised against a seven-layer, 103-marker cognitive taxonomy grounded in Naturalistic Decision Making research.

03 — Route

Context API

Agents hit the graph at decision time through a small set of well-defined endpoints. Queries return judgment ranked by confidence, recency, and relevance — at latency budgets production agents can actually afford.

04 — Audit

Decision provenance

Every agent recommendation cites only captured heuristics, with provenance to the named expert who authored it. Built for the questions model risk and audit teams ask under the EU AI Act, SR 11-7, the NAIC Model Bulletin and PRA SS1/23.

§ 04 / Timing

Why now.

Four forces, converging

01

The technology threshold lifted in 2024.

Foundation models can now parse unstructured expert reasoning, tag it against a cognitive taxonomy, and store it with provenance and confidence — at a cost low enough that the methodology is commercially viable for the first time in the forty-year history of expert-systems research.

02

"RAG over PDFs" is no longer a defensible answer to liability.

The EU AI Act is coming into force in stages, alongside SR 11-7, the NAIC Model Bulletin, and PRA SS1/23. Auditors, regulators, and trial lawyers now ask the second question: why did the model decide that? Documents alone cannot answer it.

03

Senior expertise is retiring faster than the next generation can absorb it.

Industry bodies project that up to half of today's underwriters will have retired by 2028 (NAMIC), along with a quarter of claims adjusters. Ten thousand baby boomers retire every working day in the US. The training cycle is five to ten years; the replacement pipeline runs at 60% of capacity. The window in which the senior cohort can still be observed is open right now — and substantively closed by 2030.

04

The first generation of agents shipped without judgement. The second won't.

The leading buyers in regulated work are no longer asking can the model write the report? They are asking whose judgement is in this decision, and how far can we rely on it? That is the question Tacit was built to answer. In security operations it has already arrived: AI triage now comes inside tools firms already pay for, and the firm still answers when the AI gets a call wrong.

§ 05 / Markets

Where we work.

Two products, one engine, two paths later

Underwriting · for MGAs and Lloyd's coverholders

Specialty risk, where unwritten judgement is widest.

Senior underwriters decline in seconds for reasons they cannot easily write down. Tom reads the decisions a desk's best underwriters have already made and recovers the judgement behind them: submission shape, broker history, second-order exposures. The result is a model of how the desk underwrites, owned by the MGA. Complex cyber, the towers written by hand in London rather than the SME business now priced by scan, is where the gap between the model's native capability and the underwriter's judgement is widest, and where the specialty MGAs are moving fastest on AI procurement.

We sell to MGAs and Lloyd's coverholders rather than carriers, because the pen sits closer to the person, the decision record is closer to hand, and the owner of the business is the one who feels it when an underwriter leaves.

PhaseLive
MarketUK & London first
CyberSOC · for CISOs at regulated firms

The independent judge for the AI in a security operations centre.

Tom checks the calls an AI agent has already made on the client's own alerts against what actually happened, and shows where the evidence supports letting it act alone. The vendor can't do that independently, and analysts can't do it by looking over the agent's shoulder.

Same engine as underwriting, different question: how far to trust an agent's judgement. It is read-only, runs in the client's cloud, and judges agents, not analysts. Work starts with a six-week Agent Decision Audit.

Design partnersFrom December 2026
First platformsDefender & Sentinel
Clinical decision support · horizon

The pattern recognition of senior specialists, available at every read.

A radiologist sees the malignancy before they can describe it. We encode that seeing: the gestalt cues, the comparative anchors, the second-look heuristics. The Brainoscope deployment is where the methodology is being calibrated against three expert cohorts in clinical neurotechnology — a diagnostic agent inherits the eye of the specialist, not the prose of the textbook.

Capture sessions54 · Brainoscope
Cohen's d0.92–1.82
Nuclear & industrial control · horizon

Preserving the control room as a generation of operators retires.

A reactor operator hears a drift before any alarm sounds. We capture the felt-sense of the plant — what counts as normal, what counts as worth waking the shift supervisor — before the people who hold it walk out for the last time. The expansion path through years three to five, alongside oil and gas and complex industrial process control.

PhaseHorizon
HorizonYear 3–5

— The knowledge creating company —

Every decision runs on judgement. We recover it, and show how far you can rely on it

Talk to us Read the full thesis