It’s 10 p.m., and you’re staring at your ATS. Does that resume filter count as AI? Do you need to rewrite every posting template before January 1? You scroll through the summaries of Bill 190, but the language is vague. Tomorrow, your CEO will ask if the company is compliant, and right now, you don’t have a clear answer.
January 1, 2026, is the day new rules take effect that reshape how job postings must be written. Bill 190 introduces mandatory requirements for AI disclosure in hiring and pay transparency. These changes carry compliance risk, reputational weight, and the potential to influence candidate trust from the very first interaction.
The Core Challenge: Navigating AI Hiring Compliance in Ontario
Hiring leaders already manage competing priorities: speed, accuracy, and culture alignment. Now, they must add compliance to that list. None of this arrived from nowhere. Ontario has spent years modernising employment law toward equity and fairness in hiring, and where earlier regulation focused on pay and working conditions, Bill 190 reaches further back into how employers communicate before anyone applies. The AI disclosure requirement exists because unchecked algorithms can introduce bias and erode candidate trust, and the legislature decided candidates should know when one is involved. Bill 190 requires employers to:
- Disclose whether AI is used in screening, assessing, or selecting candidates.
- Include salary or salary ranges in job postings, with restrictions on how wide those ranges can be.
- State vacancy type: whether the role is new or replacing an existing position.
- Provide timely closure: candidates must be informed of final decisions within 45 days.
On paper, these requirements are straightforward. Practically speaking, they raise questions: What counts as AI? How specific should the disclosure be? How do you frame a salary range without underselling the role or losing competitiveness?
Download the Bill 190 compliance checklist for free now.
Clarifying Grey Areas: What Counts as AI Disclosure in Hiring?
“So, does the resume keyword filter count?”
“Depends. Does it move candidates up or down the list?”
This is the kind of conversation HR leads are having in Slack and Teams right now. Bill 190 doesn’t provide an exhaustive list of what qualifies as AI disclosure in hiring, which is why employers are debating every tool, from resume filters to chatbots.
Think resume filters for software engineers, chatbots that qualify applicants, or automated video-interview scoring. They may look like small helpers, but if they change a candidate’s outcome, they need to be visible.
Bill 190 changed more than AI disclosure. The pay transparency rules, the vacancy type requirement, the ban on asking for Canadian experience and the 45-day notification deadline all landed at the same time, and each is covered in Ontario job posting requirements. This guide stays on the AI question, which is the one employers get wrong most often.
Employer Risk & Consequences of Non-Compliance
A Toronto-based SaaS firm recently posted a role without a salary range. Within 48 hours, a thread popped up on LinkedIn questioning why they were ignoring Bill 190. Comments piled up: candidates calling it tone-deaf, peers asking if the company was behind on compliance. The posting was eventually pulled, but not before the brand took a hit.
The real cost of non-compliance goes beyond potential fines. Companies risk:
- Job board rejections as platforms enforce disclosure rules.
- Candidate skepticism is voiced publicly on LinkedIn, Glassdoor, or Reddit.
- Pipeline waste occurs when candidates disengage mid-process.
Compliance protects credibility under Ontario AI hiring laws, where clear communication is quickly becoming the standard. And in the era of AI disclosure in hiring, credibility is the new currency.
A Startup Founder’s Reality
The job is critical: hiring a cloud engineer who can start yesterday. You post the role, but candidates email back with questions about salary ranges and how their resumes will be evaluated. One even asks if AI is involved. A year ago, you could’ve ignored it. Now, under Bill 190, every word in that job post is a trust signal, or a liability.
This tension captures how startup founders and CTOs feel the pressure. They can’t afford mis-hires or reputational damage, but they can’t slow down either. That’s why building compliance into every posting isn’t optional, it’s necessary.
What STACK IT Screens For
At STACK IT, compliance is built into our recruitment process. Our recruiters already apply the Bill 190 framework across both contract and permanent searches:
- AI Disclosure: We use BrightHire and Workable in our process. Candidates are told up front that these tools are AI-enhanced, but decisions remain human-led. Approved language about AI disclosure in hiring is embedded in every posting, making it clear, candid, and candidate-ready.
- Pay Transparency: Salary ranges are provided by default. If a role pays $95,000–$125,000 CAD, that range is disclosed clearly. When base pay isn’t market-leading, we highlight total compensation: bonuses, benefits, career growth, and flexibility.
- Vacancy Statements: Templates include explicit phrasing (“This role reflects an existing vacancy” or “This is a newly created position”).
- Decision Timelines: Our recruiters track closure reminders in Workable to meet the 45-day requirement.
Vertical Implications Across Industries
Compliance looks different depending on industry:
- Healthcare & Life Sciences: “We can’t afford a compliance miss when roles are tied to patient safety.”
- Finance & Insurance: “Structured compensation is part of our DNA and our postings have to reflect it.”
- SaaS & Tech Startups: “Candidates ask about equity on the first call. If it’s not in the posting, we lose them.”
- Logistics & Transportation: “Our workforce expects pay clarity up front. Without it, attrition spikes.”
Framing industry differences in the voices of hiring managers makes the nuance practical and relatable.
How Ontario compares, and where this is heading
Ontario is not an outlier. The EU AI Act requires clear communication about how AI is used in employment decisions and organises obligations by risk category. California and New York mandate salary transparency but have not yet landed on a uniform AI disclosure rule. What makes Bill 190 unusual is that it folds AI disclosure and pay transparency into one framework rather than treating them as separate regimes, which is why aligning to Ontario tends to put multinational employers ahead on both.
It is a starting point rather than a finish line. The likely direction of travel is wider definitions of what counts as AI, impact assessments for hiring tools rather than simple disclosure, and pressure toward federal or cross-provincial consistency. Employers who build the workflow now face less disruption when that arrives, and the workflow is the hard part. The disclosure sentence takes an afternoon.
Research from Glassdoor in 2024 found 72% of candidates prefer transparency about salary and hiring tools before they apply, which is the practical argument for treating this as an advantage rather than an obligation.
What disclosure changes for candidates
Candidates are already more aware of how technology shapes hiring than most employers assume, and scepticism about black-box screening tools is well documented. Disclosing AI use converts that scepticism into a trust signal, because the thing candidates object to is rarely the technology itself. It is discovering after the fact that a machine made the first call.
From the applicant side the combined effect of the new rules is less guesswork. They know the compensation before investing an evening in an application, they know a tool may screen them, and the 45-day notification rule means silence is no longer an acceptable answer. Each of those reduces the uncertainty that makes good candidates drop out of processes.
Putting AI disclosure into practice
Compliance here is a workflow problem rather than a wording problem. The disclosure line itself is one sentence. Keeping it on every posting for three years is the part that fails.
- Embed the language in templates. Approved wording goes into every job description template rather than a policy document nobody opens.
- Make the ATS fields mandatory. AI disclosure, compensation and vacancy type should be required, so a posting cannot go live without them.
- Train recruiters on the compensation conversation. A posted range changes how the first call goes.
- Track the 45-day clock in the ATS. Notification deadlines are the requirement most often missed, because nothing prompts you.
- Audit quarterly. Pull a sample of live postings and check them against the rules rather than assuming the template held.
- Keep the audit trail. Postings, interview notes, disclosures and candidate communications, retained for three full years.
Our Bill 190 compliance checklist covers the full workflow, and the job posting template has the approved language already in place.
There is a second-order risk worth understanding. Disclosing which tools screen your candidates tells a determined fraudster exactly what to defeat, which is why the wording matters as much as the fact of disclosing. We cover that trade-off in how to disclose AI without inviting fraud.
Start Strengthening Your Hiring Process
It’s still 10 p.m. and the ATS screen is still open, but this time, the templates are updated. The disclosure language is already there. Salary ranges are set, and reminders are scheduled. Tomorrow, when the CEO asks about Bill 190, she won’t have to scramble for an answer. She can say: we’re compliant, we’re consistent, and our candidates know it too. The panic is gone. The process is in place.
Hiring is about trust. Bill 190 raises the bar for how Ontario employers communicate with candidates. Those who get ahead now will turn compliance into an advantage. At STACK IT, we’ve already built it into the way we work. That’s our STACK.
Ready to align your hiring process with compliance? Book a discovery call with STACK IT to see how our recruiters apply Bill 190 standards every day.