The Big Picture:

Venture for Canada’s Student-Led AI Adoption Program is a pilot that trains students in practical AI literacy and judgment before they join Canadian small and medium-sized businesses (SMEs) for seven-week placements. After completing Anthropic’s AI Fluency course, students attend five live sessions led by AI experts, then complete additional asynchronous learning content. There, they learn to scope, evaluate, and responsibly use AI solutions that SMEs can realistically adopt.

Preparing students to bring practical AI to Canada’s small businesses

AI is changing how work gets done, but most small and medium-sized businesses don’t have the time or resources to figure out how it fits into their existing operations. Owners hear about the hype and growing expectations around AI, yet few have someone on their team who knows the difference between useful implementation and expensive distraction.

That gap is why Venture for Canada launched the Student-Led AI Adoption Program, designed to give students practical AI literacy skills and judgment before they join an SME. This way, they can help teams evaluate, design, and integrate right-sized AI solutions from day one.

The pilot program includes five live sessions with AI subject matter experts over the course of two weeks, alongside access to flexible, self-paced asynchronous learning content as they worked within real SMEs. Here’s a quick recap of what students learned and why it matters.

How the program works

Before the live sessions began, students completed Anthropic’s AI Fluency: Framework & Foundations course. It introduced them to the concept of the 4Ds (Delegation, Description, Discernment, and Diligence), a shared language for using AI effectively and responsibly.

From there, students attended five live sessions over two weeks, followed by seven weeks of self-paced asynchronous content as they worked in teams inside actual businesses. They learned the realities of a small business, how to think critically before using an AI tool, the risks and responsibilities involved with AI, how to stay in control of the work, and concluded with hands-on context engineering practice. SME partners also had their own orientation session to understand how to map their business problems and build productive relationships with their student teams.

Inside the six sessions

1. SME Reality Orientation

The first live session kicked things off by introducing students to the reality of running a small business. They explored how founders make decisions, manage competing priorities, and operate with limited time, budget, and staff. Working through a practical case study, students worked together to develop an AI solution that the hypothetical business could realistically adopt, then practiced pitching their idea to a busy founder.

Speaker spotlight: This session was led by Shadi Yazdan, a strategic AI advisor and 4x founder with 2 exits. At SY Strategy, she advises organizations on AI strategy, adoption, operational improvement, and leadership readiness. This includes helping to make decisions about where AI fits in their business, how to use it responsibly, and how to put the right governance, priorities, and accountability in place to create measurable business value.

2. Thinking Frameworks in Practice

The second live session was about building a habit: reaching for a framework before reaching for a tool. Students were introduced to the Harvard AGENT framework (Audit, Gauge, Engineer, Navigate, Track), refreshed on Anthropic’s 4Ds, and learned how to apply both to realistic intrapreneurship placement scenarios. The goal was to understand how to apply these structured frameworks to clarify the business problem first, then decide where AI can help and where human judgment is still essential before proposing a solution.

Speaker spotlight: This session was led by Brittany Perlin Danishevsky, a Forward Deployed Engineering Manager at Magical, where she partners with healthcare administrators to use AI to support and automate healthcare operations. Her career has spanned software engineering, product management, and AI-powered solutions, while her academic work at the University of Cambridge focused on AI ethics and how people interact with AI in real-world decisions.

3. Working Alongside AI, Part 1

Next, students learned what it means to treat AI as a co-thinker, rather than a replacement for their own judgment. This session covered three key risks to understand before any business data is involved: sharing confidential information, relying on hallucinated outputs, and responding to hidden or harmful instructions. It also included short scenarios in which students worked together to identify which risks were in play, how to handle them, and who owns the action items and sign-off.

Speaker spotlight: This session was led by Sophie Lemieux, an early-stage advisor, investor, and operator specializing in organizational design and AI enablement. Drawing on her background in neuroscience and more than a decade of experience with early-stage companies, she helps founders and leadership teams adopt AI in ways that are human-centered, ethical, and safe.

4. Working Alongside AI, Part 2

Part two focused on how to verify AI outputs, preserve the learning journey, and stay in control of the work. Students learned a simple verification process to check outputs for accuracy, relevancy, and completeness. They also had to consider how to use AI without giving up responsibility for the final work or opportunities for learning and decision-making. This included returning to the AGENT framework and the 4Ds to identify which parts of a task should remain human-led.

Speaker spotlight: This session was led by Liam Stock-Rabbat, the founder of Sanitized AI, a browser extension that stops sensitive data from leaking into AI tools, and founder and principal of EduRisk, a Canadian consultancy focused on human risk management. His current work sits where AI adoption meets human behaviour: shadow AI habits inside organizations, agentic AI risk, and how to train people for tools that change faster than the policy documents describing them.

5. Context Engineering

The final session was a fully hands-on workshop where students began with a vague or incomplete prompt, reviewed the weak output it produced, and improved it step by step by adding relevant context, clearer instructions, useful constraints, and a defined output format. Seeing the before-and-after helped them understand and explain the impact of thoughtful context design. The session also included the Context Engineering & Prompt Improvement Guide, which included a briefing framework, action plan, and troubleshooting steps.

Speaker spotlight: This session was led by Lawrence Quan, the founder of Kuota, a fractional growth advisory based in Toronto. Kuota embeds with early and growth-stage B2B SaaS companies as their AI-native growth team. Lawrence is a trusted advisor to founders as they build go-to-market strategies and high-taste automation for growth. He also co-leads the Toronto chapter of Operators Guild, a global community of Series A-D operators.

6. SME Partner Organization Orientation

While students were learning practical AI skills and judgment, SMEs also participated in a separate orientation session, where they learned how to direct students toward high-impact business problems while maintaining clear guardrails around data security and supervision. This session included problem scoping, student supervision guidelines, AI safety parameters, and a Q&A section to address any questions before starting with their respective student teams.

Speaker spotlight: This session was led by Diana Khater, Managing Director at Dianara Advisory, a Toronto-based consulting firm that helps small and mid-sized businesses grow more efficiently by fixing the commercial and operational gaps that quietly hold companies back. Whether it’s a process that’s stalled, systems that don’t talk to each other, or a team that’s busy but not converting, Diana works alongside business owners to find the fix and put it into practice.

Early impact

Right now, over 100 students are participating in the Student-Led AI Adoption program; live sessions concluded on October 1st while the cohort will end on November 20th. They aren’t just learning how to use AI, but also how to form the right mindset in applying AI responsibly and effectively. Students will leave with an understanding of how to apply these foundational skills to real-world challenges: what does this business actually need, what can AI tools realistically support, and where should a person stay in the loop?

This pilot would not have been possible without Shadi Yazdan, Brittany Perlin Danishevsky, Sophie Lemieux, Liam Stock-Rabbat, Lawrence Quan, and Diana Khater. Each brought real-world experience that turned AI literacy from an abstract idea into practical, usable skills. We’re grateful for their time and commitment, and encourage VFC’s community to follow their work.

Once the pilot program concludes in November, we’ll be gathering feedback from students to shape future programming. Learn more about VFC’s other student programs here.

FAQ

What is Venture for Canada’s Student-Led AI Adoption Program?

It’s a pilot program that prepares students to help Canadian SMEs adopt AI in practical, realistic ways. Students first complete Anthropic’s AI Fluency course. They then attend five live sessions with AI subject matter experts over two weeks, supported by self-paced online content. Afterward, they apply what they’ve learned during a seven-week placement with an SME, who have also completed their own orientation session. The goal is for students to be useful from day one, helping businesses evaluate, design, and integrate AI solutions that fit their time, budget, and staff capacity.

What foundational AI skills do students need before joining an SME?

Students need to understand how small businesses actually operate, including founders’ competing priorities and limited resources, so they can propose AI solutions the business can realistically adopt. They should use structured thinking frameworks, such as the Harvard AGENT framework and Anthropic’s 4Ds, to clarify a problem before choosing a tool and understand when to use human-led judgment. They also need to recognize key AI risks: exposing confidential data, relying on hallucinated outputs, and falling for prompt injection. Finally, they should know how to verify AI outputs for accuracy, relevance, and completeness, and how to write clear, context-rich prompts that produce reliable results.