Learn › Agent Harness, the big picture › Stop 2

Stop 2 of 9 · Page

Autonomous agents

What autonomous agents are, the five levels of autonomy, why businesses want them, and the catch.

What it is

An autonomous agent is an agent that nobody is watching step by step. It starts on an event or a schedule, works for minutes or hours, and acts by itself inside limits that someone set in advance.

Autonomy is not a fourth kind of application. It is a second dial. The kind (stop 1) says who decides the next step. Autonomy says how much a person checks.

0 · Answera person reads itand decides1 · Suggesta person takesthe action2 · Act with OKa person approveseach important step3 · Act in limitsa person reviewsafterwards4 · Unattendeda person sets goalsand reads the reportmore freedom for the AI, less checking by a person
Autonomy is a dial with five positions, not a switch. Most business value today sits at levels 2 and 3.
LevelWhat the AI doesWhat the person doesExample
0 · AnswerAnswers one questionReads it and decidesSummarise this email
1 · SuggestProposes an actionTakes the actionDrafts a reply for a claims handler to send
2 · Act with approvalActs, but asks before anything importantApproves or rejectsAsks before making a refund
3 · Act within limitsActs by itself inside a budget and a set of rulesReviews afterwardsSettles small claims and logs every step
4 · Run unattendedStarts itself, runs for hours, recovers from problemsSets the goals and limits, reads the reportHandles incidents overnight

Why businesses want them

Each benefit below comes with the number you would measure to prove it. A benefit you cannot measure is a claim, not a benefit.

BenefitIn an insurance claims teamHow you would prove it
Work does not wait for office hoursA claim reported at 11 pm is checked and acknowledged at 11 pmTime to first response
Faster from start to finishSmall, simple claims settled in minutes, not daysTime from report to settlement
Spikes without hiringA storm triples the claims for a week; the queue does not tripleSize of the backlog
People do the judgement workHandlers take the disputed and the large claims; the agent takes the routine onesShare of staff time spent on exceptions
The same rules every timeEvery claim is checked against the same policy wording, and every step is written downRule breaks found in audit
Lower cost for each piece of workThe cost of a settled claim drops, most of all when simple rules run before the model is calledCost for each settled claim

Worth it when the work is…

  • high in volume and repetitive
  • governed by clear rules
  • easy to undo, or cheap to get wrong
  • measurable: you can tell a good result from a bad one

Not worth it yet when the work is…

  • rare, or different every time
  • impossible to undo and high in stakes
  • impossible to judge afterwards
  • cheaper with a simple rule or a workflow

The catch

With nobody watching, there is no person to catch a mistake. Each problem below has an answer, and the answers together are what this course calls the harness.

ProblemWhat it looks likeWhere we meet it
It does not know when to stopIt loops all night, or says the job is done halfwayStop 4: the limit in the loop
It can be trickedA sentence hidden in a document becomes an orderStop 5: the no-guardrails simulation
It can reach too muchIt reads or changes things the person asking has no right toStop 5, second story
Small mistakes add upA wrong reading at step 3 becomes a wrong payment at step 9Stop 6: reins that pull it back early
Nobody can see what it didA customer complains and there is no record of whyStop 8: the See family

Next: enough talk. Stop 3 puts ordinary code and an agent side by side, doing the same job.