${...} expressions send their contents to an LLM and replace the expression with the response. They’re how you embed reasoning, classification, extraction, and generation directly in step config — no Code step required.
Don’t confuse this with the Generate tab in a smart value field, which uses AI once at design time to write a deterministic
{{ ... }} JSONata expression. ${...} (the AI tab) runs an LLM call on every workflow execution. See Generating expressions with AI.Basic syntax
${...} is the prompt sent to the model. {{...}} interpolations are resolved against ctx before the prompt is sent.
How they evaluate
- JSONata is resolved first —
{{...}}becomes literal values - The resulting prompt is sent to the configured AI model (Azure OpenAI by default)
- The response replaces the
${...}expression
gpt-4o-mini). For higher-stakes uses, override the model in the step’s settings.
Pure vs interpolated
Same rule as JSONata: a pure AI expression preserves the model’s response type when possible.87, the field becomes the number 87. If the response is "87 - the lead is qualified", the field becomes the string "87 - the lead is qualified" (no parsing).
For numeric and boolean outputs, be specific in the prompt: “respond with only the number”, “respond with exactly ‘true’ or ‘false’”.
Prompt patterns
Classification
Extraction
Summarization
Boolean judgment
Generation
When to use AI vs JSONata vs Code
Model selection
By default AI expressions use a small fast model. To override, configure the workflow’s AI model in Settings → AI:
Cost and latency rise with model size — only step up when small models fail your eval.
Reliability patterns
Constrain the response format
The biggest source of AI expression failures is loose response format. Always be explicit: ✅ “Respond with only one word: yes or no.” ❌ “Is this qualified?” ✅ “Respond with a JSON object with keys ‘name’ and ‘email’.” ❌ “Extract the contact info.”Use few-shot for tricky classifications
Guard with downstream validation
After an AI expression, validate with JSONata or a Code step:Cost and latency
Each AI expression is one LLM call. They add up:- A workflow with 5 AI expressions costs roughly 5× a single LLM call in time and money
- For agent tools (sync, on the call), aim for ≤2 AI expressions per workflow to stay under the latency budget
- For async workflows, cost matters more than latency — use AI freely
Limits & gotchas
- AI expressions are non-deterministic. The same input can produce slightly different outputs. Don’t use them where determinism matters (use JSONata or Code).
- Long prompts cost more and may time out. Keep transcript / context interpolations short or summarize first.
- The AI doesn’t see
ctx— only what you interpolate into the prompt with{{...}}. If your prompt needs upstream data, interpolate it explicitly. - Don’t put secrets in prompts. They’re sent to the LLM provider and may be logged.