> ## Documentation Index
> Fetch the complete documentation index at: https://docs.trinigence.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Limits of Automation

> What ATI can automate reliably - and where human intent and judgment still matter.

## The core limitation

ATI can translate intent into strategy logic.

But it cannot replace:

* your risk tolerance
* your market assumptions
* your decision to deploy live

Automation is powerful, but it has limits.

***

## What ATI automates well

ATI is strongest at:

* translating indicator logic into deterministic rules
* completing missing structural fields (when possible)
* generating backtest-ready strategies
* providing clarity and iteration suggestions

***

## Where automation breaks down

### Vague intent

Examples:

* “trade smart”
* “avoid chop”
* “enter on strong moves”

These require definitions.

### Market prediction

ATI does not forecast price direction.

### Unspecified execution details

Slippage, partial fills, and exchange quirks vary in real life.

### Over-optimization

ATI can help iterate, but cannot protect you from chasing noise unless you follow guardrails.

<Card title="Overfitting & guardrails" icon="shield" href="/iteration-optimization/overfitting-and-guardrails" horizontal>
  Protect against curve fitting.
</Card>

***

## Practical takeaway

Use ATI to:

* translate and test fast
* iterate safely
* reduce mistakes in structure and risk

Do not use ATI to:

* assume profitability
* ignore execution realities
* deploy without understanding the trade behavior

<Note>
  ATI reduces translation friction.\
  It does not remove responsibility.
</Note>