AI in hiring only delivers when the data underneath it is complete, and the data only stays complete when everyone actually uses the system. This page lays out how we think about that, and what it means for building AI you can trust with a hiring decision.
Most ATSs fit one shape of hiring well. The teams that shape doesn't fit end up working around the system instead of in it, keeping their own spreadsheets, running a separate tool for the deskless roles, moving the senior hires along over email.
By the time AI shows up, it's working from whatever made it into the system, which tends to be the easy half of the story and not the part where the hard judgment calls happened. People treat that as as an AI problem. It's a data problem, and the data problem comes down to a simpler question: does everyone actually use the system?
Getting there takes three things at once. The system has to be flexible enough to handle every kind of hiring you do, give you the controls to ensure consistency and compliance when it's needed, and stay easy enough that recruiters, hiring managers, and candidates reach for it instead of around it.
When all three hold, people use it. When people use it, the record stays complete, and a complete record is what lets AI work on the actual hiring story. Most ATSs manage one or two of the three. Pinpoint was built for all three at once.
Pinpoint is already moving from an ATS with AI features toward one where AI enables you to work wherever you work: inside Pinpoint or inside the assistants you use in your day-to-day work like Microsoft Copilot, Claude, Gemini, and ChatGPT.
From there it gets more active, with Pinpoint handling more in the background, surfacing the candidates that need attention and drafting the next step for you to approve without prompting.
Further out, you'll be able to hand it whole sections of your hiring process, with approval checkpoints along the way. What doesn't change is that as AI takes on more of the work, the decision making remains yours.
AI reads a resume and summarizes an interview, it drafts the follow-up message, and then a person decides what to do.
A human is required at every decision point. AI supports recruiters and hiring managers through the work faster. It doesn't decide who gets hired.
The most legally exposed pattern in AI hiring is behavioral inference: a system that learns which candidates you've hired before, and then reproduces those patterns.
Pinpoint's AI doesn't work that way. It works from the input you give it in that moment and the data sources you point it at. It doesn't study how similar candidates moved through your funnel, and it doesn't quietly form a view of the kind of person you tend to hire. What you ask for is what you get back.
Every AI-derived suggestion in Pinpoint is transparent and auditable. When a candidate gets a match score, you can open it up and see why they scored the way they did against the criteria you set. There's no black box. If we can't make an AI output explainable, we don't build it.
Your data is never used to train, fine-tune, or improve any model. AI features are opt-in, and you decide what's switched on.
The rules around AI in hiring are moving quickly, and they're landing on the employer and the vendor. The specifics differ, but most regulations are asking for similar themes: transparency, a human in the loop, and the ability to show how a decision was reached.
Pinpoint was built with that in mind rather than patched to meet it after the fact. A person decides at every step, suggestions are explainable and auditable, and the AI works from what you tell it rather than from patterns in your hiring history.
For the candidate-facing side, you get notifications, disclosures, and you choose what to share and when.
Pinpoint is among the first ATSs to achieve ISO 42001 certification, recognizing our commitment to responsible, ethical, and transparent AI.
All uses of AI within Pinpoint are lawfully formed and respect both copyright and other 3rd party and privacy rights.
Yes, you can choose which of our AI features you want to adopt.
Yes, access to configurable AI features can be restricted based on user permissions. In these cases, there are distinct permissions for configuring vs viewing the output of the AI feature.
No. AI surfaces the insight and your people make the call. A human is required at every decision point, and the AI never decides who advances or who gets rejected. We believe there should always be a human in the loop.
No, customer data is not used for training our models.
Use AI for the parts of hiring that involve reading and drafting, and keep a person on every decision. In practice that means letting AI summarize resumes and interviews, draft candidate messages, and surface a shortlist, then having a recruiter or hiring manager review and decide. AI only works well on complete hiring data, so the whole team needs to work in one system. Summarizing resumes and interview notes, drafting candidate communication for a human to approve, surfacing strong-fit candidates for a recruiter to review, and keeping a person accountable for every hiring decision are just some of the ways AI can be used in hiring.
AI can both introduce and reduce bias in hiring, depending on how it's built. The biggest risk is behavioral inference, where a system learns who you've hired before and repeats the pattern. Pinpoint's AI is built to avoid that. It works from the input you give it, and every output is explainable, so you can always check the reasoning.
No. It takes the repetitive reading and drafting off recruiters' plates so they can spend more time on judgment and on people. In Pinpoint, every AI output is something a person acts on, not a decision the system makes on its own.
In most places, yes, though it's increasingly regulated. Laws like New York City's Local Law 144 and the EU AI Act set requirements around bias auditing, candidate notice, and human oversight. Requirements vary by region, so check what applies where you hire.
