Ottawa Will Co-Fund 10,000 AI Work Placements
The federal government announced on September 25 that it will provide up to $162 million over five years, beginning in 2026–2027, for Mitacs to support 10,000 co-funded work placements. Students, recent graduates and post-doctoral fellows go into Canadian businesses to work on AI adoption and innovation. For the large share of small businesses that cannot justify a hire and do not want a consulting engagement, this is the most relevant AI funding announcement of the year.
What was announced
The announcement came from the Minister of Artificial Intelligence and Digital Innovation, who is also the minister responsible for the Federal Economic Development Agency for Southern Ontario. The money funds Mitacs AI Advantage programming, which runs through two complementary streams.
| Stream | Focus |
|---|---|
| ADOPT | AI adoption. Helping businesses test and deploy AI solutions |
| AI+X | AI innovation. Applying AI to business problems and new technologies |
ADOPT is the one most small businesses will care about. It links industry partners with academic researchers and talent to help organisations of all sizes test and deploy AI, which is a fair description of what a ten-person firm actually needs. AI+X leans toward research collaborations, with agriculture and health care named among the disciplines.
It sits under the AI for All strategy, which the government says will support up to 90,000 AI-related jobs and work placements for young Canadians by 2031. The stated diagnosis behind it is one this site has been making for months: small and medium businesses can see the potential and lack the skills, capacity and support to implement it.
Read the headline number carefully
Three things worth holding onto before anyone builds a plan on it.
Up to $162 million. Over five years, beginning in 2026–2027. The phrase is doing work, and it is a ceiling rather than a committed total.
Ten thousand placements over five years. Roughly 2,000 a year, across the whole country and every sector. Meaningful, and not unlimited.
Co-funded is not free. The business contributes a share. How much depends on the stream, the institution and the current terms, all of which change, so confirm with Mitacs or a university co-op office rather than budgeting from a summary.
None of that makes it less worth doing. It makes the difference between arriving with a realistic expectation and arriving having told your board about free AI talent.
Why this route fits AI work
Most of what a small business needs at this stage is not building. It is finding out. Which of three tools actually handles your documents. Whether the saving on a task is real once verification is priced in. What the process currently is, written down.
That work is time-boxed, produces an artefact, and does not justify a permanent hire. It is also exactly what a structured student term is shaped for, and it is the work a consulting engagement tends to price out of reach for a firm under twenty people.
The businesses that get value from this will be the ones that arrive with a defined problem. We set out what a placement-sized project looks like, and the four things that waste the term, in co-op students and how small firms get AI help.
What to do before you apply
Pick the problem. One measurable thing that is slow or expensive now. Not a technology, a task. The exercise in process mapping produces this in an afternoon, and the timings you record become the evidence in the application.
Name the supervisor. A person, with an hour a week in the calendar. Programmes ask, and the placement fails without it regardless of what the form says.
Contact a co-op office as well as Mitacs. Universities and colleges near you run placement programmes continuously and know which funding a given project fits. They also know the timing, and academic terms do not start when you decide you are ready.
Decide the access rules now. Own account, only the systems the project needs, nothing irreversible. That is the same rule as for any new person, and it is much easier to set before someone starts than to retrofit in week six.
The broader set of Canadian programmes, including BDC LIFT and the regional initiatives, is in Canadian AI companies and what counts and our guide to AI grants and funding. If you need help writing the application itself, the split between what to draft and what to write yourself is in grant writing with AI.
Frequently Asked Questions
What did the government announce?
On September 25, 2026, the Minister of Artificial Intelligence and Digital Innovation announced up to $162 million over five years, beginning in 2026–2027, for Mitacs to support 10,000 co-funded work placement opportunities. The placements are for post-secondary students, recent graduates and post-doctoral fellows working with Canadian businesses on AI adoption and innovation, delivered through Mitacs AI Advantage programming.
What are the two streams?
ADOPT supports placements focused on AI adoption, linking industry partners with academic researchers and talent to help businesses of all sizes test and deploy AI solutions. AI+X supports placements focused on AI innovation, co-funding research placements that apply AI to business problems and new technologies across disciplines including agriculture and health care. ADOPT is the one most small businesses will be looking at.
Is the placement free for the business?
No. These are co-funded placements, meaning the business contributes a share of the cost and the programme covers the rest. The specific contribution depends on the stream, the institution and the current programme terms, and those change, so confirm the split with Mitacs or a university co-op office rather than budgeting from any published summary including this one.
How many placements will actually be available?
10,000 over five years, so roughly 2,000 a year spread across the whole country and all sectors. That is meaningful and it is not unlimited, and the headline figure is described as up to $162 million rather than a committed total. Treat it as a real programme worth applying to rather than a guarantee of a placement whenever you want one.
Who is this actually good for?
A business that has identified a specific AI project, can name someone to supervise it, and has been unable to justify either a hire or a consulting engagement. It suits evaluation and documentation work particularly well. It suits a business hoping someone will arrive and work out what the AI project should be considerably less well, because that is not something a student can determine for a company they have never seen.
Arrive with the project already defined
We help Canadian businesses turn a vague sense that AI should help into a scoped, measurable project a placement can actually finish.
Related Articles
Billable Hours When AI Does Part of the Work
Will AI Take My Job? What the Numbers Say
What If a Platform Wrongly Flags Your Work as AI?
Ajan leads the ChatGPT.ca team: 200+ custom GPT builds and automation projects for 50+ businesses across 20+ industries. Based in Markham, Ontario. PIPEDA-compliant solutions.