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

> ## Agent Instructions
> This is the Journeyfront Help Center for hiring teams. Journeyfront is an applicant tracking system (ATS). Use the Help Center's terms: job req, Candidate Profile (the person), Candidate Application (one person on one job req), hiring class. Link to the article you used in your answer.

# How Journeyfront uses AI

> Where AI helps in Journeyfront, and the guardrails that keep your team in charge of every hiring decision.

> [!NOTE]
>
> **Availability:** Each AI feature needs to be turned on for your account, and some also need a permission or a setting on each job req. Ask your Customer Success Manager. AI agents use AI credits.

When you're hiring at volume, a lot of the work is reading: hundreds of resumes, hours of recorded answers, and follow-ups with candidates who stalled halfway through. Journeyfront uses AI to take on that first read and the routine follow-up, so your team can spend its time on the candidates and decisions that need a person. AI gives you a head start; your team makes the call.

## Where you'll find AI

> [card] [AI resume scoring](https://help.journeyfront.com/scoring/ai-resume-scoring.md)
>
> Scores each resume against the attributes you set for the job, explains every score and highlights the lines behind it.

> [card] [AI-scored questions](https://help.journeyfront.com/scoring/ai-scored-questions.md)
>
> Scores video, audio and written screening answers from 1 to 5 against your scoring criteria, with its reasoning.

> [card] [AI agents](https://help.journeyfront.com/workflows/ai-agents.md)
>
> Follow up with candidates who stalled, by email, text or WhatsApp, with your team approving messages when you choose.

> [card] [AI Connections](#ai-connections)
>
> Connect your own AI assistant to read your Journeyfront data.

| Feature | What it does | What it never does | Who's in control |
| - | - | - | - |
| **AI resume scoring** | Gives each resume a score per attribute, with reasoning and highlighted evidence | Hire, reject or move a candidate | Turned on per job req. Your team can edit any score, and edits are marked. |
| **AI-scored questions** | Scores free-form answers against your criteria, with reasoning | Hire, reject or move a candidate | Turned on per question. Permissions decide who can turn it on, run it and edit AI scores. |
| **AI agents** | Writes and sends short follow-up messages to stalled candidates | Move, reject, score or assign anyone; message hired, rejected or withdrawn candidates; text people who haven't opted in | You choose the candidates, the channels and whether each message needs approval. Every action is logged with its reasoning. |
| **AI Connections** | Lets an assistant you connect read your Journeyfront data | Change anything in Journeyfront | Each user connects their own assistant and can revoke it at any time. |

## Our principles

### People make the decisions

AI in Journeyfront informs decisions; it doesn't make them. AI scores are one input alongside your assessments, interviews and your team's judgment. AI doesn't hire, reject or move candidates, and AI agents can only send messages.

Your automation rules are yours to design. If a rule acts on scores, those scores can include AI scores, so build rules that send borderline candidates to a person for review rather than rejecting them outright. See [Automation rules](https://help.journeyfront.com/workflows/automation-rules.md).

### People can review and change every AI result

* **Every AI score shows its reasoning,** and resume scores highlight the evidence in the resume.
* **Your team can change any AI score.** For resume scores, edited attributes are marked as edited by a person.
* **AI agents can wait for approval.** For each channel, choose whether an agent sends on its own, asks when it's unsure, or always asks first.

### AI only sees what the task needs

Each AI feature works from the information it needs for its job: the resume and the job's attributes, a candidate's answer and your scoring criteria, or a candidate's progress and message history. AI agents work only from hiring workflow data, never from protected characteristics like age, gender or race.

### Your data isn't used to train AI models

Your candidates' and employees' data isn't used to train the AI models that Journeyfront's features use.

### Security

AI features run inside the same security program as the rest of Journeyfront, which is SOC 2 compliant.

## AI Connections

With **AI Connections**, you can connect an AI assistant that supports MCP (Model Context Protocol) to your Journeyfront account, and ask it about your candidates and applications in plain language. The assistant reads your data; it can't change anything in Journeyfront.

**1. Open AI Connections**

Click your name at the top right, then **Personal Settings**, then **AI Connections**.

**2. Copy the connection URL**

Click **Copy** next to **Connection URL**, and follow **How to set this up** for your assistant.

**3. Approve access**

Your assistant sends you to Journeyfront to sign in and approve access. Once approved, it appears under **Active connections**.

![AI Connections in Personal Settings, with the Connection URL and its Copy button, How to set this up, and one assistant under Active connections with a Revoke button](https://mintcdn.com/journeyfront/61dBaB1hIRAsZpUl/images/getting-started/ai/gen-ai-connections.png?fit=max\&auto=format\&n=61dBaB1hIRAsZpUl\&q=85\&s=82ca66d8718f441f8f82d7b1e82057d0)

The assistant only sees data from the account you're working in, and only what you could see yourself. To disconnect it, click **Revoke** next to it under **Active connections**.

To have your assistant answer how-to questions from this Help Center as well, see [Use the Help Center with your AI assistant](https://help.journeyfront.com/getting-started/use-docs-with-ai.md).

AI Connections needs to be turned on for your account, and your security role needs the *Connect to MCP* permission. What a connected assistant does with your data is governed by your agreement with that assistant's provider, so check your company's AI policy before connecting one.

## Automation that isn't AI

Some features do work for you without AI:

* **Automation rules** follow exactly the rules your team sets. See [Automation rules](https://help.journeyfront.com/workflows/automation-rules.md).
* **Fraud flags** come from automated checks on each application and give it a risk level for your team to review. See [Fraud flags](https://help.journeyfront.com/candidates/fraud-flags.md).
* **Predictive models** are statistical models Journeyfront builds from your own employees' outcomes. See [Predictive models](https://help.journeyfront.com/scoring/predictive-models.md).

## FAQ

> [accordion] Is AI deciding who gets hired?
>
> No. AI scores and agent messages support your team. People review the results, can change them, and make every hiring decision.

> [accordion] Can we use Journeyfront without AI?
>
> Yes. Each AI feature is turned on for your account only when you ask, and AI resume scoring and AI-scored questions are also turned on per job req or per question. Everything else in Journeyfront works without them.

> [accordion] Can we see why the AI gave a score?
>
> Yes. Every AI score comes with its reasoning, and resume scores highlight the lines of the resume behind them.

> [accordion] Can the AI get it wrong?
>
> Yes, which is why people stay in the loop. AI only judges what's in front of it: it can't confirm a claim is true, and it can't credit experience a candidate left out. Treat a low AI score as a reason to look closer, not a verdict, and change any score you disagree with.

> [accordion] Is our data used to train AI models?
>
> No. Your candidates' and employees' data isn't used to train AI models.

> [accordion] Do we need to tell candidates we use AI?
>
> Rules about using AI in hiring differ by location, and some require notice to candidates. Check with your legal team about the places you hire.

> [accordion] How do we turn on an AI feature?
>
> Ask your Customer Success Manager to turn it on for your account. Then give the right people the permissions in **Settings**, then **Security**, and turn the feature on where you want it: on a job req for AI resume scoring and AI agents, or on a question for AI scoring. See [Security roles and permissions](https://help.journeyfront.com/account-setup/security-roles.md).

## Related

> [card] [AI resume scoring](https://help.journeyfront.com/scoring/ai-resume-scoring.md)
>
> Score resumes against the job's attributes.

> [card] [AI-scored questions](https://help.journeyfront.com/scoring/ai-scored-questions.md)
>
> Score video, audio and written answers.

> [card] [AI agents](https://help.journeyfront.com/workflows/ai-agents.md)
>
> Follow up with stalled candidates.

> [card] [How scoring works](https://help.journeyfront.com/scoring/how-scoring-works.md)
>
> How AI scores fit into a candidate's overall score.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.