Video summary
Maxim Jean-Louis, the president and CEO of Contact North in Ontario, Canada, delivered a thought-provoking presentation at the Eden conference in Porto, focusing on the evolving relationship between students and artificial intelligence. His talk, titled "When Students Trust AI More Than People," centers on a critical mental health risk emerging in educational settings. He observed that students who feel anxious, lonely, or depressed may find solace in AI chatbots because these machines listen, respond, and make sense of their feelings without judgment. Over time, this dynamic can shift the student's trust from human beings to machines, creating a situation where the digital tool becomes a safer emotional refuge than real people.
The core of Jean-Louis's argument revolves around the concept of "cognitive offloading," where students increasingly rely on AI not just to support their thinking but to perform thinking tasks entirely on their behalf. He distinguishes between an AI acting as a "learning partner" that helps users understand concepts and a "thinking partner" that formulates thoughts for them, noting that the latter represents a fundamentally different and potentially dangerous relationship. This shift is particularly concerning when students use AI to organize scattered or emotional thoughts, effectively outsourcing their internal processing. For vulnerable individuals, the chatbot may cease to feel like a mere tool and instead become the primary source of reassurance, guidance, and companionship, which poses significant risks to their psychological well-being.
To address these challenges, Jean-Louis proposed four actionable strategies for educators and institutions. First, they should openly ask students about their AI usage during wellness conversations to understand the extent of reliance without policing behavior. Second, faculty and counselors must be vigilant in recognizing when AI is replacing necessary human connections rather than enhancing learning. Third, AI literacy education needs to expand beyond technical skills to include honest discussions about trust, dependency, emotional attachment, and the inherent limitations of these systems. Finally, institutions must prioritize making human mentors, peers, and advisors more accessible, emphasizing that stronger human connection is the essential antidote to the loneliness that drives students toward over-reliance on technology.
Jean-Louis concluded by urging the Eden community to approach this issue with humility and a willingness to share experiences, acknowledging that there are no simple final answers yet. He highlighted that while not every student using AI is in trouble, enough cases have emerged to warrant immediate attention from those who work closely with students. By comparing notes on observed struggles and fostering open dialogue about these changing dynamics, the educational community can better understand and mitigate the risks associated with deepening AI dependency. Ultimately, his message was a call to action to ensure that technology serves as a bridge to human connection rather than a substitute for it, protecting the mental health and holistic development of the student population.
Read the full video transcript
It's Maxim Jean Louui who is president
and chief executive officer of contact
north in Ontario, Canada. That
organization is a good friend of Eden
and works with Ontario's colleges,
universities, indigenous institutes and
other public education and training
providers.
Its work focuses on improving access to
learning, supporting innovation through
applied research and technology and
sharing knowledge in Canada and
internationally.
And today, Maxim will share some
thoughts, in fact rather questions with
us. Maxim, the floor is yours.
>> Thank you. You've noticed how relaxed
Bim is this morning.
Yesterday I was having dinner with him
while Belgium was drilling Egypt
and he was sweating but today he's
feeling much better.
Let me first apologize to you because
what I didn't want to do this morning
was to repeat the things that have been
said yesterday.
So I decided to adjust my presentation
somewhat because that's the challenge
when you are the speaker on the second
day.
So my title now is
when students trust AI more than people.
I'll repeat that. When students trust AI
more than people,
the mental health risk
Let me start with a mental health risk.
Student is anxious,
lonely
or depress.
They open an AI chatbot.
It answers.
It listens.
It makes sense of them.
And slowly it becomes easier to trust
the machine.
than to trust another person.
What do we do when that begins to affect
their mental health?
Need to ask ourself that question.
Let's reflect together
over the next 10 minutes.
Last night during the wonderful Eden
Gala dinner cruise on the Duro River
here in Porto, I was speaking with my
longtime colleague and friend John
Baker, founder and executive chairperson
of D2L,
one of Canada's largest learning
management system companies.
John mentioned a study by I'm going to
murder their names, Chang Lee, Young
Huey, and Linda Sarah Hijan titled
the cognitive impact
of Chad DPT in higher education.
a systematic review of critical and
creative outcomes,
creative thinking outcomes.
Right there on the boat, John did what
he always did.
Email literally the study to me.
So before most of Porto this morning
I was reading that study over coffee
only to be interrupted by Nicole Johnson
who was going out to jo.
The study reviewed 67 research papers on
Chad DPD in higher education.
For me, it came down to
two words.
Cognitive
offloading.
Students using AI
not simply to support their thinking,
but increasingly to perform thinking
task on their behalf.
That phrase kind of sat with me for a
while.
Yesterday
after the very informative and
comprehensive keynote by Professor Luis
Paloise,
I asked him about something on his very
first flight. He had described AI as a
learning partner.
I asked him whether he was in fact
holding back, whether learning partners
stopped short of something stronger.
Maybe AI is becoming a thinking partner.
A learning partner helps us understand.
A thinking partner helps us formulate
the thought itself. There is a
difference.
It's a totally different relationship.
Then there is my wife Le who some of you
met yesterday.
She's a retired teacher whose pastime is
stock investing.
Doesn't like Elon Musk at all.
She says something that made this whole
issue less abstract to me a few days
ago.
He says, "The reason I talk to my AI
rather than write to it is when I talk,
I can get all my thoughts out at the
same time. Even if it is mumbo jumbo,
if my thoughts are all over the place,
here I still figure out the main idea
and give me an answer."
This is the point.
You can be scattered,
unclear,
emotional,
all over the place, and the AI still
answers.
It organizes you and me. It makes sense
of you and me. That is useful.
It may be powerful in ways we have not
yet fully understood.
for a student
who is lonely,
anxious,
isolated
or vulnerable.
The chatbot may not feel like a tool.
It may feel like the one place where
they are understood.
This is where my worry begins.
Not with AI writing an essay,
not with AI solving a problem,
but AI becoming the place a student
turns for reassurance,
guidance,
companionship.
So let's ask this in a blunt way.
What happens when students begin
trusting AI more than people?
Colleagues, it is happening.
What happens when the tool starts to
feel like the safest woodenship in their
life?
And what happens when that begins to
affect their mental health?
I'm not suggesting every student who
spends significant time with AI is in
trouble,
but many of us in this room work closely
with students.
I'd be surprised
if some of you have not already seen
something that made you pause.
A student leaning too heavily on AI.
A student turning to AI for emotional
support.
A student whose use of AI seems to be
much more than learning.
I think something is starting to show
up. Not everywhere,
not every student, but enough that we,
the Eden community, should pay attention
to.
So what do we do?
What do we do?
Four possible options.
First, ask.
When we talk to students about wellness,
let's ask about AI use, not to police
them, just to understand.
Second
notice
faculties,
advisors,
counselors
and student support staff need to
recognize when AI may be replacing human
connection
instead of supporting learning.
Third,
let's teach the risk
honestly.
AI literacy which we all talk about
should include trust,
dependency,
emotional attachment
and the limits of those systems.
Fourth and last,
make people easier to reach for the
student.
We all know the words mentoring,
peer support,
advising,
community
because loneliness
is part of the problem. The answer
cannot only
be better technology.
It has to be stronger human connection.
Let's compare notes.
As I said yesterday,
most wise approach to all of this is
humility.
I do not have a final answer, but I do
think we need to start comparing notes.
If you've seen students struggling and I
know that you have some of you with
this, I do hope you will share those
experiences during this second wonderful
day of the Eden Conference
because if something is changing,
we'll understand it faster
together than alone.
Thank you.
[applause]