A robot arm on a test stand moving one joint at a time through a slow self-check

Technology

Adaptive runtime autonomy.

Models give machines capability. Runtime autonomy gives them judgment on the job: the ability to learn, decide and stay within the rules while the work is happening.

01 / The shift

From trained once to learning always.

Trained, frozen, shipped

  • Learns only when retrained
  • Every request pays for the full model
  • Rules checked after the fact
  • A new case waits for the next release
  • What it learned cannot be undone

Adaptive at runtime

  • Learns from every decision, live
  • Serves what it knows itself
  • Rules enforced at the moment of decision
  • A new case learned from one example
  • Any lesson removed exactly, on request
A gantry robot drilling an aircraft wing panel

02 / Principles

Six principles behind every decision.

  1. 01

    Learn in the moment.

    A single example, correction or demonstration becomes usable skill on the very next decision.

  2. 02

    Act on confidence.

    The runtime acts when it is sure, asks when it is not, and turns every answer into experience.

  3. 03

    Hold every rule.

    Policies, limits and keep-out zones are part of the decision itself.

  4. 04

    Keep the record.

    Every lesson and every action is written to a tamper-evident record.

  5. 05

    Reverse anything.

    Any lesson can be removed exactly.

  6. 06

    Give back to the model.

    Field experience can be exported as a training set that makes the next base model stronger.

03 / Where it sits

Beside your model, not instead of it.

The machine sends what it sees to the runtime gate, which commits, defers to the policy or model, or refuses, under the operator's rules, and writes every decision to the record.OperatorRules and limitsMachineRobot, vehicle or agentSeesActsMA Reflex / MA DuplexRuntime gateCommitDeferRefuseLearns · remembers · decidesWhen unsureLearnedPolicy or modelThe one you already runRecordEvery decision
  1. 01

    Connect

    Place the runtime in front of your model or beside your machine's policy.

  2. 02

    Work

    It handles what it is sure of and hands off the rest.

  3. 03

    Learn

    Every handed-off answer, correction and demonstration becomes experience.

  4. 04

    Compound

    The share it handles grows and costs fall.

The autonomy system

04 / Economics

Intelligence that compounds.

Each answer the runtime learns is one it never has to buy again.

A mobile robot carrying a stack of totes along a warehouse lane and turning into the next aisle

89%

Fewer model calls on repeated work, with MA Duplex

2.4×

Fewer policy calls per episode with MA Reflex, on a simulation benchmark

An uncrewed combat aircraft in level flight over mountain ridges

See runtime autonomy on your own work.