Canadian AI Advancement Institute

Canadian AI Advancement Institute Our mission is to ensure that every Canadian has the opportunity to learn, adapt, and succeed with AI.

We accomplish this through institutional and industry partnerships focused on delivering AI education and training.

08/21/2026

A model that performs strongly in a test environment will eventually be deployed in the real world, and the physical world isn't always agreeable; surfaces are uneven, sensors are noisy, and people wave their hands in front of the camera because sometimes they choose to be silly.

"Those virtual AIs, at that time, it was kind of easy mode, because you could do anything in a simulator. You can kind of just say I'm going to go to this location, but in real life there's friction, there's slippage on the floor that makes it so as much as you would try to go there, it doesn't always work out the way you planned." (Angelica Lim)

This gap can catch researchers off guard as they move virtual AI deployments from test environments to physical AI systems.

08/20/2026

The next technology shock is already on the calendar. We just don't know the date.

"I think when quantum breaks, we're going to have exactly the same situation we had in 2022, and everyone's like, oh, what is this now?" (Timothy King)

How do you think society will react when quantum arrives on the global stage?

08/19/2026

Many stakeholders say AI is moving faster than we can govern; few share what governing it would actually involve.

For Atif, "anytime AI has to enter the organization, it has to go through the governance vetting process, and that's where the algorithmic impact assessment comes in. Right now we're at about 96 to about 100 plus questions that we will vet any AI model." (Atif Malik)

A hundred questions sounds heavy until you compare it to the cost of finding out after deployment that a system was never assessed for the people it affects. It also changes who has to be in the room, because a vetting process that only engineering can answer is not a governance process.

08/18/2026

Researchers have started using the terms AI-associated delusions or AI psychosis for instances when someone engages in unhealthily long conversations with an AI model.

"These AI models do not have that grounding to reality. They also don't sleep, compared to a friend which would be like, okay, now let's just go sleep. Our discussion's gone on way too long." [Angelica Lim]

A friend eventually tells you to go to bed or revisit the discussion another time. A chatbot rarely does.

Nothing in the exchange pushes back, tires, or ends; the ordinary corrections a human relationship supplies without anyone having to ask for them. That makes this a design problem as much as a clinical one, and it lands hardest on the people using these tools most heavily and least visibly.

What safeguards do you think should be in place to prevent this?

08/14/2026

"If we actually explain to people how LLMs work and they could see the machinery in action, there wouldn't be any of this weird anthropomorphizing going on where we're treating it like it's a person, because it isn't. It doesn't think. It cannot perceive the world. It's a wonderful statistical trick that can automate certain processes" (Timothy King).

Timothy’s perspective is an interesting and realistic approach. To stop people from seeing AI as a ‘person’, we will need to invest in AI literacy from K-12 to seniors.

How do you think we can better educate stakeholders about the reality of LLMs?

08/13/2026

"A strong strategy is one that is relatively all-encompassing... cover[ing topics such as] ethics, security, machine learning, model training, prompt engineering, practical applications, customization" (Tristan Taylor). Prompt engineering is typically where most organizations start, as it can quickly produce a visible result.

Have you or your team received any AI-related training? If so, what was your experience, and how has it helped you?

08/12/2026

Building a strong AI literacy strategy is the key to ensuring safe and ethical technology implementation in organizations of any size.

As Atif mentioned, a successful strategy focuses on enabling people rather than scaring them away from the technology. This involves providing guardrails, safe sandboxes for experimentation, and establishing a dedicated committee led by an internal champion, such as a Chief Data Officer.

What strategies is your organization using to build AI literacy and safely explore new use cases?

08/11/2026

If educational institutions are "only focusing on surveillance instead of education on AI tools, we're really missing the point. As educational institutions, it's not the right focus. Students are here to learn. So how do we... create that culture of trust back in to be able to help students use AI right?" (Tiffany MacLennan)

We’re seeing that typically, a student whose AI use is monitored learns to avoid being caught, and not necessarily how to use a tool well. For an institution to be a leader, it's best to take the approach that focuses on student success.

08/11/2026

"We've been seeing a lot of institutions adopt kind of general training models on AI literacy... They give people a common language to be able to talk about these things, maybe in ways that they wouldn't be able to otherwise." (Tiffany MacLennan)

General AI training puts everyone on the same footing and provides a shared understanding. Without that, governance conversations can stall on definitions, pilots get assessed against assumptions nobody stated out loud, and whoever holds the strongest opinion ends up setting the vocabulary for everyone else. Shared language is not the whole of AI literacy, but it is the part that has to come first.

08/08/2026

As Timothy King states, utilizing lesson content straight from a Large Language Model can remove the educator's personal touch and input from the process. This disconnection can create significant issues on multiple fundamental levels.

Do you think AI-generated lesson content can hinder classroom connection and have a negative effect on students? Why or why not?

Address

Langley, BC

Website

Alerts

Be the first to know and let us send you an email when Canadian AI Advancement Institute posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.

Contact The School

Send a message to Canadian AI Advancement Institute:

Shortcuts

Share

Category