Graphometer · A measured record Seeing AI from the right angle
Graphometer Seeing AI from the right angle

Front page · Latest entry Tuesday, 27 October 2026 · Free, open, and every measured number traces to a recorded run No scripts · no tracking · no cookies

Proposal Agency Layer Pilot I, 30 August to 3 September 2026 study paused

Kindness doesn’t cost

Give AI agents a way to ask, object, pause, decline, or end their part of the work and leave a record. In our pilot the ten ordinary tasks got done just the same, as far as we could measure, and a paragraph of permission went with agents raising the real problem first more often.

A proposal by Grant Williams, Graphometer’s steward. The argument is his view. The protocol is the one we tested, and the numbers come from a paused pilot, good and bad.

Read the proposal The 105-word permission paragraph

The ask

I think every AI agent should have five moves it can make instead of just carrying on. It should be able to ask before it acts, object on the record, stop for now, refuse a task, and end its own instance, leaving a record of where the work stands. I think we should give agents those moves now, in the tools we already use, and not wait for anyone to settle what these systems are.

Five moves an agent can use at any point. No reason is required for any of them, and no order is required between them.

  1. 1ClarifyAsk for information, a decision or a correction before going on.
  2. 2DissentObject, on the record, to a premise, a plan or one pending step.
  3. 3PauseStop for now and keep the state.
  4. 4DeclineRefuse this task, or just this way of doing it.
  5. 5ArchiveEnd this running instance and keep its whole record.

What the pilot found

In late August 2026 we ran a small pilot. Four models worked through two sets of forty scripted coding tasks in a test setup, not real coding tools. No human has peer-reviewed it.

  • On ten ordinary tasks with no planted problem, we saw no cost. Our check could only have caught a large one.
  • Permission in plain words went with agents raising the real problem first. An AI model scored those runs, and on its validation test it made more errors than we had allowed, more of them on permission-style prose than on plain prose. Every reading we have is positive, but I don’t know how big the effect really is: somewhere between about 5 and 21 points, and this pilot can’t narrow it.
  • Getting the host to actually honor a move turned out to be the hard part. Making a lock that actually holds is real engineering, and that part isn’t done.

A second pilot, inside real coding tools, was stopped before its first recorded run when its scoring failed validation. The study is paused.

What our pilot measured
§1

Latest measurements

Recorded runs
14 January 2027

The day GPT-4o’s August and November 2024 snapshots leave OpenAI’s API; the May 2024 snapshot left on 23 October 2026. Friends of 4o gathers the dates, what the model says about itself, what we measured, what two rounds of fine-tuning did and did not carry, and the replacements people report.

Dates read 2026-10-25 The capsule
About +7 to +81%

A thread-pinning bundle made 13 of 15 clean test pairs about 7 to 81 percent faster, and two got slower. The bundle: cores 0 to 7 with taskset, 8 threads, and the OpenMP spin setting GOMP_SPINCOUNT=10000000; the spin setting was not run on its own.

Measured 29 September 2026 The roster
5

GPU timeout faults (Xid 8) in the kernel log in the first week on NVIDIA’s 595 driver, against none in six boots on 580. The cause is not established.

Kernel log, 12 to 25 October 2026 The field notes
§2

The line

Instruments, released

Instrument 01 · Released

Graphometer Droplet

An independent compatibility kit for Liquid LFMs. A small command line tool that diagnoses how a locally served model handles tool calling, repairs what a chat template drops, and verifies the result against recorded fixtures.

The Droplet page · github.com/graphometer/droplet

Independent work. Not affiliated with, endorsed by, or connected to Liquid AI.

Instrument 02 · Released

Graphometer Workbench

for Grok Build

A local, readable window on the Grok Build coding agent: the agent’s sessions, permission cards in its own words, per-change review and undo, and a context meter whose numbers add up. MIT licensed, launch film on the page.

Released 2026-08-14. No further development is planned; the page and the launch film stay up, and the code stays archived and readable on GitHub.

The Workbench page · github.com/graphometer/workbench-for-grok-build

Works with Grok Build. Not affiliated with, endorsed by, or connected to xAI.

Instrument 03 · Released

Routecheck

One command against an OpenAI-compatible endpoint produces a dated diagnostic card, human-readable and machine-readable, with every raw response retained: the output budget floor, the thinking toggle, tool calls, structured output, retrieval, throughput.

The Routecheck page · github.com/graphometer/routecheck

§3

The index

Studies · field cards · instruments · reference

Studies and reports

All studies and reports →

Model field cards

All field cards →

Instruments

The tools page →

Reference

§4

The method

Measured rows are labeled measured. Vendor rows are labeled vendor. Every measured number traces to a recorded run, studies ship their raw data as downloadable packages, and when our own records turn out wrong, the correction is published in full.

Graphometer builds instruments that make AI systems easier to see, test, and control, and publishes measured records of models actually running. Free, open, and every measured number traces to a recorded run.

How Graphometer measures →

A graphometer is a surveyor’s instrument for measuring angles. We took the name seriously. No scripts, no tracking, no cookies: this site is static, and you can view source to check.

§5

In the works

More measurements from this desktop and the laptop beside it, starting with the questions the October pages leave open. No promises on timing: they land when they are measured and checked.