Artificial intelligence is no longer just a technology-sector issue. It’s transforming how public services are organized and delivered, including settlement services for immigrants and newcomers in Canada.
Settlement services are often the first bridge to housing, employment, language training, health care, legal information, education and community life for newcomers. In Canada, the settlement system is vast: in 2024–25, more than 550 partner organizations helped slightly more than 690,000 newcomers outside Québec alone.
Digital tools have already entered settlement work. The important question we need to answer is how AI should be used: as a tool that strengthens human services, or as a shortcut that risks replacing professional care with automated answers?
As an immigration policy researcher, the answer is clear to me: AI can help settlement agencies work faster and reach more people, but it should support workers, not replace them.
Read more:
Are algorithms unfairly screening out immigrant job applications?
Where AI can help
A well-designed AI system could answer routine questions at any time of day. It could help newcomers find English or French classes, employment workshops, food banks, legal clinics, housing information or community events. It could also support staff by organizing referrals, sending reminders and reducing repetitive administrative tasks.
This fits the federal government’s AI Strategy for the Federal Public Service 2025–2027, which emphasizes human-centred, responsible and secure AI adoption. The strategy presents AI as a way to improve services and help workers focus on more complex tasks, not as a replacement for human judgment.
Settlement work is a good example of why that distinction matters. Newcomers rarely need only one piece of information. A person may need a job but also language training, child care, credential recognition, tenant information and help understanding health care. These needs are connected, and they often require trust.
Research on human-centred AI in Canada’s settlement sector argues that AI could be useful because newcomers need timely information and service providers face heavy information demands. But it also calls for responsible tools designed with service providers, governments and communities, not generic chatbots dropped into complex social services.
THE CANADIAN PRESS/Nathan Denette
Information, navigation and training
AI may be most useful in three areas: information, navigation and training.
For information, AI could help newcomers quickly find basic answers: where to register for language classes, how to prepare for the citizenship test, where to get help with a résumé or how to contact a settlement agency.
For navigation, AI could guide newcomers toward the right type of support. A person may not know whether they need a settlement worker, language assessor, employment counsellor, legal clinic or regulated immigration consultant. AI can help identify the right door, but it should not pretend to be the expert behind that door.
For training, AI could support resumé preparation, mock interviews, job-search workshops, pronunciation practice, workplace vocabulary and informal language learning. This could help newcomers with irregular work schedules or those living far from in-person services.
The digital divide is still real
Research on digital settlement services has found that online and hybrid delivery can improve access to information, employment support and language training. It also found barriers related to broadband, digital literacy, language proficiency, privacy and access to technology.
Having a smartphone is not the same as having stable internet, a private space, enough data, a laptop or the confidence to complete forms online. A newcomer may be comfortable using WhatsApp but still struggle with government portals, digital document uploads or virtual classes.
Accessibility is also a major concern. Statistics Canada reported that in 2024, 45 per cent of people with disabilities or long-term conditions faced barriers in online activities because of their condition. This is a warning for any settlement system that assumes digital services are automatically accessible.
AI cannot replace immigration experts
The highest-risk area is legal and immigration advice. AI tools may provide general information, but they should not act as immigration consultants.
In Canada, paid immigration or citizenship advice must come from authorized representatives, including licensed immigration consultants, lawyers, eligible Ontario paralegals or Quebéc notaries.
A wrong answer about a language class may be inconvenient, but a wrong answer about immigration status, refugee protection, family sponsorship or legal rights can be life-changing.
AI also cannot understand the full emotional, legal and social context of a person’s life. It may miss trauma, fear, abuse, exploitation or urgent risk. Human escalation is therefore essential. AI should help identify when a case is simple and when it must be referred to a qualified person.
(Getty/Unsplash+)
AI will change hiring
Some agencies may see AI mainly as a way to save money. But responsible AI is not free. Organizations need people who can protect privacy, update information, monitor errors, test accessibility and decide when a case must move from a chatbot to a qualified worker.
AI may not eliminate the need for settlement workers. It may change the kinds of skills agencies need. Future teams may include settlement workers, immigration experts, language instructors, employment counsellors, data specialists, cybersecurity staff and accessibility advisors.
Settlement work is relational. Newcomers often trust workers who understand their language, culture, migration experience and local community. A chatbot cannot replace that trust, but it can support it when connected to real people.
A hybrid future
Canada should not reject AI in settlement services. The technology has real potential. But governments and settlement agencies should avoid building a digital-only system.
The better model is hybrid: AI for routine information, translation support, navigation and basic triage; humans for complex cases, legal referrals, trauma-informed care, accessibility needs and emotional support.
The future of settlement services will likely be more digital. But success should not be measured by how many workers are replaced. It should be measured by whether newcomers receive faster, safer and more inclusive support.
AI can open the door. Human expertise must remain inside.
The post “Canada’s settlement services are going digital. AI should support workers, not replace them” by Hamed Kazemzadeh, Adjunct Assistant Professor, Peace and Conflict, University of Calgary; L’Université d’Ottawa/University of Ottawa was published on 10/08/2026 by theconversation.com