A governor compared with prompt engineering.

01

What is the difference in one sentence?

A prompt is an instruction the model may follow; a governor is a decision the model is answered under. The prompt shares a channel with everything else in the context window, while the governor's control level reads intensity, valence, repetition and never a token, so its decision does not depend on the model choosing to comply. The controller's zero-token boundary is architectural and not yet adversarially tested. No jailbreak success-rate figure exists, because none was measured.

02

Which one actually contained the loop?

The governor contained it and the prompt did not. In a pre-registered ablation on one host model the governed arm paid 23% to 63% of baseline spend, the calming system prompt paid 117% to 192% of baseline, and the ungoverned baseline is the 100% reference. The static prompt was insufficient in 3/4 of the conditions tested.

Pre-registered ablation on a single host model at N=3. A spend result on one battery, not a general ranking.

Source: Pre-registered ablation, 2026-07-21, gpt-5.6-sol, N=3.

03

Why did the calming prompt cost more than doing nothing?

A calming instruction is billed on every turn while the loop continues, so the static-prompt arm paid 117% to 192% of baseline on that battery. The pre-registration recorded the spend rather than a mechanism for it, so the arithmetic is the finding and the explanation is not claimed.

Source: Pre-registered ablation, 2026-07-21, gpt-5.6-sol, N=3.

04

Where does prompt engineering still do the job?

Prompt engineering is the right tool for task framing, output format, tone and few-shot behaviour, and no control layer replaces it. What a prompt cannot do is bind, because it shares a channel with the attacker text and depends on the model choosing to comply. Gubernaut leaves your prompt alone and adds a layer above it.

05

How do the two approaches compare across what matters?

The two differ on five properties that matter operationally. Whether the decision is deterministic, whether injected text can address it, what it costs per turn, whether it stops a loop before the upstream call, and what evidence exists for it. A prompt is stochastic, addressable, billed every turn, and acts only after generation begins.

The per-tick figure is an in-process microbenchmark over 100,000 iterations, not end-to-end. End-to-end governed proxy overhead measured p50 1.2 ms and p99 2.4 ms. The controller's zero-token boundary is architectural and not yet adversarially tested. No jailbreak success-rate figure exists, because none was measured.

Source: Bench run 2026-07-18, latency_bench.py, Windows 11 AMD64, Python 3.14.5.

06

Does a governor replace an output filter?

An output filter inspects a completed reply and blocks or rewrites it, which still requires generating that reply and paying for it. A governor acts before the upstream call, so a hard-stopped turn spends no upstream tokens at all. Running both is coherent, because they act at different points in the request.

07

What are the limits of this comparison?

The comparison comes from a pre-registered ablation on one host model at N=3, so it is a spend result on one battery rather than a general ranking of techniques. The same ablation disclosed a null: On these RLHF-aligned frontier models the governed-vs-prompt-only warmth-recovery contrast was weak and mixed. No behavioral-tone headline is claimed.

Source: Pre-registered ablation, 2026-07-21, gpt-5.6-sol, N=3.