Often, yes, and even where the law is silent, you should. Some rules require notice when AI helps make hiring decisions, others when a bot could mislead someone about being human, and disclosure is the safest default everywhere.
There is no single US federal rule on this. Instead, a patchwork of state, city, and EU rules applies depending on where candidates are and what the AI does. Below is what we could verify as of October 2026.
Do Laws Require Telling Candidates About AI in Hiring?
It depends on the tool and the location. Two different ideas are at work:
- Bot disclosure: telling someone they are talking to a machine, not a person.
- Decision-tool notice: telling candidates that software will be used to evaluate them.
Recruiting agents that screen, schedule, or answer questions can touch both.
What Do the Main Rules Say?
California: bot disclosure (SB 1001)
California makes it unlawful to use a bot to mislead a person online about its artificial identity in order to incentivize a commercial purchase or sale, or to influence an election. A person is not liable if they disclose that it is a bot, and the disclosure must be clear, conspicuous, and reasonably designed to inform people. Whether a recruiting chatbot falls under the “commercial” trigger is a legal question for counsel, but the clear-and-conspicuous standard is a useful model. California also has separate rules on human review of AI hiring decisions; see our guide to California AI hiring human review rules.
New York City: Local Law 144
Employers and employment agencies must give notice to candidates before using an automated employment decision tool. The city’s Department of Consumer and Worker Protection (DCWP) says the notice must be given 10 business days before the tool is used. The tool must also have had a bias audit within one year, with audit information publicly available. DCWP began enforcing in July 2023. For background on audit math, see the four-fifths rule explainer.
Maryland: facial recognition in interviews
Maryland bars employers from using certain facial recognition services during an applicant’s interview unless the applicant consents by signing a waiver. It took effect October 1, 2020.
Colorado: a moving target
Colorado delayed its AI Act to June 30, 2026, and lawmakers then took up a 2026 replacement. Reports indicate a new law that emphasizes notice and disclosure for automated decisions in areas like employment, with a later start date. Confirm the current text and dates with counsel before relying on any of this.
European Union: AI Act Article 50
For candidates in the EU, Article 50(1) requires providers to design AI systems that interact directly with people so that those people are informed they are dealing with an AI system, unless that is obvious to a reasonably well-informed person. The European Commission says the transparency rules take effect in August 2026, and describes the recent “digital omnibus” timeline shifts as applying to high-risk rules, not transparency. The omnibus regulation (EU 2026/1744) entered into force July 27, 2026.
How Do the Rules Compare?
| Rule | Who it covers | What it requires | Status (Oct 2026) |
|---|---|---|---|
| California SB 1001 | Anyone using a bot online with Californians | Clear, conspicuous bot disclosure to avoid liability for misleading about artificial identity | In effect; verify on leginfo |
| NYC Local Law 144 | Employers and agencies using automated employment decision tools in NYC | Advance candidate notice (10 business days per DCWP) and annual bias audit | Enforced since July 2023 |
| Maryland (HB 1202) | Employers using facial recognition in interviews | Signed applicant consent waiver | In effect since Oct 1, 2020 |
| Colorado AI Act | Developers and deployers of high-risk AI | Originally delayed to June 30, 2026; replacement law reported | In flux; verify |
| EU AI Act, Art. 50(1) | Providers of AI systems that interact with people | Inform people they are interacting with AI unless obvious | Applies from August 2026 |
What Is the Best Practice for Disclosing AI to Candidates?
Disclose early, plainly, and every time, and always offer a path to a person. Do not wait for a law to demand it. Good disclosure is also good candidate experience, and it supports the principles in responsible AI hiring.
Sample language: “Hi, I’m an AI assistant for [Company]. I can answer questions and help you schedule an interview. A member of our recruiting team reviews hiring decisions. You can ask for a person at any time.”
Best-practice checklist
- Identify the AI in the first message, before any questions.
- Use plain words such as “AI assistant,” not vague terms like “virtual helper.”
- State what the AI does and does not do, including whether it evaluates candidates.
- Keep a human accountable for hiring decisions; see human-in-the-loop recruiting.
- Offer an easy way to reach a person and an alternative to AI steps.
- Give any legally required advance notice (for example NYC) in a separate, timely message, not buried in terms.
- Get explicit consent for biometric or facial analysis where required.
- Log when and how each disclosure was shown.
- Review the rules by candidate location at least quarterly.
- Audit tools for bias and publish results where required.
Joveo’s AI agents operate with human oversight, which reflects this approach.
Frequently Asked Questions
Do I have to tell candidates they are chatting with a bot?
No single US federal law requires it, but California’s bot law, NYC’s notice rule, and the EU AI Act can apply depending on the tool and location. Disclosing in every case is the lowest-risk approach.
Does disclosing that a chatbot is AI hurt candidate engagement?
We have no verified data on this. Clear disclosure paired with a quick route to a human is widely seen as the better candidate experience.
Does NYC Local Law 144 apply to chatbots?
It applies to automated employment decision tools used to screen candidates for employment or promotion in New York City. A chatbot that only answers FAQs may not qualify, but one that scores or ranks candidates could. Ask counsel.
When do the EU transparency rules apply?
The Commission states the AI Act transparency rules take effect in August 2026. Check current guidance, since implementation guidelines and codes of practice were still being finalized.
Is the Colorado AI Act in effect?
Its original date was delayed to June 30, 2026, and a 2026 replacement was reportedly enacted with a later effective date. Confirm current status before relying on it.
Where can I learn more about agentic recruiting?
Start with Joveo’s guide to what agentic recruiting is.
Not legal advice. This article is general information, not legal advice. Laws change quickly and vary by jurisdiction; consult qualified counsel.
















