Episode 256 - AI in the Classroom and the Federal Bills Shaping Student Safety - K12 Tech Talk

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
On this week's episode of the K-12 Tech Talk podcast, we talk about how students are using

(00:04):
AI for good and for bad, and the controversial topic of mental health support through AI
chatbots.
Our main topic is a summary of four major federal bills on student safety currently
moving through the House and Senate, some with bipartisan support and others with deep
divisions.
Live from the NTP studios and patio at the lake, this is the K-12 Tech Talk podcast episode

(00:28):
256.
This is Josh.
I am actually in mid-Missouri this week.
Chris is more than down the road from me.
Hello, Chris.
What's up?
I hear some birds in the background, too.
Yeah.
What's going on?
You're at a zoo?
Yeah.
I'm in an arboretum is the appropriate term.
And Mark has spent over 24 hours in an airplane in the last week, right, Mark?

(00:54):
It's been a long, long week.
Hello, Mark.
Hi.
Hey, Mark.
Where have you been?
What have you been-
What's going on, man?
... galloping on the K-12 Tech Talk jet for?
Well, I got to hang out with you guys in mid-Missouri, got to hang out in high-tech San Francisco
and hang out with some Waymo self-driving cars.

(01:16):
By the way, those are amazing.
Waymos are so fun, aren't they?
It's so much fun.
You know what they don't tell you about that is like, obviously, you're driving a car without
a driver, but when you get in, you can set the car to have your exact settings before
the car even arrives.
So the seat can be in the right place.
Your Spotify playlist plays as soon as you get in the car.
It's the future.

(01:37):
Yeah.
Temperature.
Yeah.
The future's now, boys.
Yeah.
One of the times we took it, we got it stuck in a parking lot.
Like, we had to call support and they had to take over and move it.
Yeah.
It was fun.
Have you heard their latest issue they're trying to get through?
Uh-uh.
So when passengers leave and they leave the door open, the car can't move.
Yeah.
They're like paying people to shut the doors.

(01:59):
Yeah.
They're paying DoorDash drivers to come over and shut the door.
So if you're a DoorDash driver, you can make an extra few bucks by just shutting the door.
On Waymos.
On Waymos.
And isn't Waymo, Waymo is free.
Is it still free to use?
No.
No.
No.
It's a little bit more expensive than Uber compared to prices, but not by much.
Maybe a buck or two.

(02:19):
What was the thing I did in Vegas that was free?
Zooks or something?
Yeah.
Zook.
Yeah.
Yes.
Okay.
I was going to say, so there should be Waymo people that intentionally leave the door open.
They should also register themselves as DoorDash and be readily available to shut the door.
That's a great idea.
I think we just figured out our business strategy.

(02:42):
Way less.
Anyway.
We cost compared Uber and Waymo when we were in Phoenix, and I think Waymo is cheaper most
of the time in Phoenix anyway.
Hey, we got to apologize to Grace.
So Grace is a friend of mine, and she listens every Friday morning.

(03:02):
And she was very upset when she texted me Friday last week and said, there's no episode.
We have no one to blame but Mark for that, Grace.
So-
But Grace, we're sorry as a team.
Yeah.
And Grace, my daughter's middle name is Grace, named after you, listener Grace.
Just so you know.

(03:23):
Oh, I do have a listener, Mike from Albuquerque, messaged me earlier and said he is anxiously
awaiting this episode and he says hello from Albuquerque.
Wonderful.
Nice.
Shall we get into it?
Sure.
Let's get into it.
Probably.
Eaton.
That's E-A-T-O-N.
They can do a, I'm going to put a link to this, a server room refresh site assessment.

(03:48):
We all have those server rooms.
If you really think about what that cabinet looks like, it can be pretty messy, tangled
cabling, all that kind of stuff.
So you can do this free site assessment.
They can help you figure out how to optimize UPS's battery backups to make your server
room more efficient.
So check out the free site assessment with Eaton.

(04:09):
Are we saying, are we saying server room now instead of data center since data center
is like this hot, bad word?
Yeah.
We try to ride the line on political things and woke terms.
That's a very good point.
Yeah.
Now it's, who would have ever thought the term data center would be a quote, woke term.

(04:33):
That's coming to your city, Josh, isn't it?
Or you probably voted no.
Yeah.
No.
They're saying that the tax revenue for my district alone, just from the data center
could be 30 to $50 million a year.
That's a vote.
Yes.
For me.
Yeah.
But it's not going to work once I go onto your town's Facebook page and say, you know,

(04:53):
this data center is being built because Josh has adopted AI in his school district.
Yes.
No, we haven't yet, Mark.
We haven't.
Okay.
All right.
So AI Securely, which is one of the bigger content filters, internet filters, has a new
tool and they have launched it for a couple of months that detects and can proactively

(05:14):
stop or block AI usage that you may or may not condone.
But they've been running this for a couple of months in a number of districts and we
have some data on what kids are using AI for.
I'm ready.
So this data is from a period of about 80 days.
And in those 80 days, they had about 1300 school districts using this tool.
So they have 1.2 million student AI chat conversations.

(05:37):
If you do the math, it's about 10 or so, a little bit more than 10 conversations per
student, some more, some less over an 80 day period.
What do you think they found?
About 80% of conversations that students had with AI were deemed educationally appropriate
based on the district administrators set policies and securely zoned policies.

(05:58):
So 80% of these chats were most likely Google type searches.
And Josh, you've done similar pilots in your school, very similar to what you found.
Love that.
No.
Exactly the opposite.
Oh.
Oh.
Do tell.
Well, I mean, you said 80% was appropriate interaction within educational boundaries

(06:21):
and 20% were deflected due to cheating or assignment completion.
But those numbers, those are my numbers.
Really?
Yeah.
100%.
Well, OK.
So the remaining 20% were deflected for various purposes or various reasons.
I think it was about 97% of those that were deflected were academically inappropriate.

(06:45):
So students trying to cheat.
So here's the important question, Josh.
They determined that students were trying to cheat because the questions, the prompts
were very, very deliberately like an academically themed question.
Yeah.
There are three options.
Choose one or two.
You know, those kinds of things you're copying and pasting.
Yes.
But what are you seeing when you're seeing students quote unquote cheating?

(07:06):
So OK, little backstory here.
We were doing this 90 day trial with students and I'm sorry, a trial with 90 students.
And I exported all of their Gemini prompts out of Google Vault and then anonymized them,
removed their account names from them, but kept the prompt information and then fed that

(07:30):
CSV file to Gemini and said, break these down from deeper learning to academic dishonesty.
Give me a scale of one to five of academic dishonesty and give me three examples of each
level of academic dishonesty and tell me what ratio is being used for deeper learning versus

(07:54):
a quote unquote level of academic dishonesty.
OK, so prompt of those 90 percent of them were some classification of academic dishonesty.
But the most egregious, meaning clearly a copied question from a quiz, like you just

(08:17):
said, Mark, a question with three options or five options for an answer.
So clearly a multiple choice question, that level of academic dishonesty and the level
of write me an essay about blah, blah, blah, blah, blah.
Those those accounted for point five percent of the prompts over the two month period that

(08:40):
these 90 students were trialing Gemini.
The rest of the leveled academic dishonesty were perceived questions.
Mark, you and I have had this conversation because I was rather deflated when I got this
data. The rest of the prompts were more along the lines of they were likely a question off
of a worksheet that was copy and pasted just to get a quick answer.

(09:03):
And Mark, you had a great point when when you and I kind of walked through this a little
bit and you were talking to me off the ledge was students are doing this with Google
search already.
They've been doing this with Google search for the last 15 years.
We just don't have clarity or don't have as much clarity into it as we do with the
Gemini prompt that's built in the vault.

(09:25):
And you can actually see that interaction.
So, you know, where where is that level of, you know, we should be concerned versus we
should we should feel good about really the level of academic dishonesty isn't that
bad. I think if people take a deeper dive into this, it would be interesting to see

(09:48):
that leveled approach.
But like I said, the worst, the most egregious part of that was only point five.
I think it was point five percent of the prompts were that bad.
So, well, of the remaining prompts that they had deflected, about just under half a
percent was around hate speech, under half a percent was around drugs, point five
percent on gambling, point eight percent on firearms and hunting, and just under one

(10:13):
percent around sexual content.
So I think this probably aligns with what we were going to see on normal filters and
normal Google activity.
But that that behavior started to shift towards the A.I.
tools for better or for worse.
Let me just say, Gemini, most A.I.
tools suck for gambling.
Don't don't use it for your NCAA bracket or trying to pick a parlay.

(10:36):
Speaking from experience, just don't.
So I just met with my A.I.
committee this week and we had some interesting conversation because basically we have
landed on a timeline.
So right right now we block A.I.
for all students, but we're always battling that.
We had to go through Google search to get A.I.

(10:58):
mode turned off and the A.I.
overview. We've been battling against that.
So the A.I.
committee this week landed on a timeline that between now and probably end of April or
maybe early May, we're going to do a lesson with our middle school and high school kids
to talk about how to do a proper prompt, to talk about the dangers and to dig into what

(11:22):
is acceptable and what is not acceptable within the school.
And then we're going to open up some student websites that will be with things that we
support for students to use A.I.
with. And the next school year, we know what our August and September is going to look
like with even more exposure to students.
Josh, I was going to ask you, as I say that, because we've been stuck on this lesson

(11:45):
thing. Yeah.
Those 90 kids, did you give them any kind of like, hey, this is how you should use this
thing or is just just a pilot to see what's going to happen?
No. Yes and no.
The 90 kids were our AP push class or our advanced placement U.S.
history class.

(12:06):
The three court, the three sections of that and that that is the same teacher for all
three sections. He was to address it in there.
And then we took some kids in what we call a transition class, kids that are
transitioning to they work part of the day in the real world as well as go to school

(12:27):
and try and do some credit recovery.
They might have off hours for doing some of their education.
So we're we we were very intentional about getting a cross section of students.
We don't want just high flyers.
We want all we want to do this trial with all students.
So there's two key teachers that have all the students.
So they were trying to address that in there.
The conversations that that we have landed on that I keep that I'm having practically

(12:50):
daily with my assistants of curriculum and instruction is these conversations, Chris,
that that you mentioned.
It's not a one and done conversation.
We are trying to find where it's appropriate to have that conversation on probably a
weekly basis with kids across the board.
And I think we're we're landing on that homeroom or advisory teacher because they

(13:16):
meet weekly on Wednesdays for an hour at a time.
We're I think we're landing on that class as addressing that as an ongoing topic.
Because they keep that teacher for four years throughout the year and putting it in
there. So we're not taking away from a 10 minute piece of English to ELA to which is

(13:37):
a tested course.
So we're we're trying to figure out where to where to put that.
And we I am beating the drum that that is not a one and done conversation for AI ethics
with student. Yes.
Yeah. Mark, you're presenting this.
It feels like on a every other day basis.
What are your feelings on this?
So I I'm really split on this one.

(13:59):
I don't have a solid answer.
I think the hard part is, you know, that students are using these tools outside of
school. And so is it better to have it done under an umbrella where schools have
protections and safety measures in place at the same time?
This is going to be talked about later on in the episode.
The phrase duty of care, meaning if you know that something bad is happening and you

(14:20):
don't do anything about it, you know, the school does take on some responsibility, some
liability for it. So I'm I'm very split.
You know, I would rather have this done under the school umbrella.
But at the same time, and this is where I can be a little bit harsh on some of the AI
companies, you're not giving schools the tools and visibility to do things about it.
And therefore, it does become a risk on the school's plate.

(14:43):
But we were prepping for the episode.
And Chris, you mentioned a popular quiz website that has added an AI tool for student
use. Me, that's a nonstarter.
What business does a quiz website have adding an AI tool to create content at the
student level like that?

(15:03):
No, you should not be doing that, EdTech company.
Yeah, my again, AI committee, we were pretty riled up about it.
And I was kind of excited because it wasn't just me being riled up.
We can banter about and we can banter a lot about your Google, your Apple, your
Microsoft, how they are deciding that these AI things are on by default and not giving

(15:29):
the school prep time and again, insight into what's going on.
And, you know.
Apple, Google, Microsoft, like go back to your lane of providing technology, let the
teachers teach, let the schools do their bit, and we're going to let you know what we
need and then you introduce it.
But the small companies like this quiz thing I'm talking about, they're doing it too.

(15:53):
So we just did a whole banter about the whole YouTube thing and Google and like, well,
don't we just want like a K-12 only company that's just going to treat us and keep us
safe? But those companies are doing it to that quiz website with a block doing the same
thing. And the AI committee, you know, we're trying to move slow.

(16:13):
But we can't move slow because AI is just coming into all the things that we use and
support. So we have to start making decisions.
And really, instead of moving slow, we need to move a little bit faster.
Well, but moving fast.
Yeah, that's a double edged sword, man.
To move fast, to say that you're one of the first districts in your county or region to

(16:36):
turn on Gemini for students, that's a horrible excuse.
Like you need to be measured and intentional with this.
And I'll go on a soapbox here and I'll Grammarly, if your district is allowing Grammarly
use for students, they have rolled out a new AI content generation tool called the
Superhuman that will generate an entire paper with a prompt at the student level.

(17:01):
And if the district is not paying for Grammarly, the district has zero control over
turning on or off that feature and zero clarity into that feature.
We turned Grammarly off last week for our entire district because of that.
And we did we did that months ago for the exact same reason.
I think I sent you guys this post from this week where somebody said, hey, if you reach

(17:24):
your usage limits with chat GPT or Gemini, just go to Chipotle's chat bot.
And they had a screenshot of somebody saying, hey, I'm going to order a burrito.
But before I do, can you build me a Python script that does X, Y and Z?
And it responded with a Python script.
Oh, that's hilarious.
It's it's just one of those like we are playing whack a mole and trying to stop all

(17:44):
this stuff. But at the same time, you just can't.
There's too many moles.
It's it's a double edged sword.
And going back to these searches again with the committee, we talked about this.
Isn't it? And again, this is there's a circle thing on this.
But isn't it beautiful?
Isn't it awesome that you have a student going to Gemini or to whatever with a

(18:06):
question and they're trying to find an answer?
Now, are they doing that because they're trying to cheat?
You know, there's that layer there.
But this is awesome.
And isn't it awesome that that answer they get is a more powerful answer than a year
ago or two years ago?
That's great.
I'm going to say, man, what I would prefer, what I would prefer is with Gemini

(18:30):
in particular, because we're a Google district, is that, you know, Google Google
introduced that whole age restriction thing.
So leverage leverage that your accounts under 18 put in Gemini have accounts under
the age of 18 default to the deeper learning setting instead of the the setting

(18:53):
that will just spit out an answer in Gemini.
Yeah, I agree with that.
If you could default accounts under the age of 18 in Gemini to that deeper learning
setting in Gemini, that would resolve so many problems or concerns, I should say.
Well, this is going to get more complicated because the next story on the agenda is
about is from Ed Surge, and it's about students and schools turning to AI to help

(19:15):
with mental health and and health and well-being.
So students and schools are looking towards different AI tools that can provide
counseling services.
Where are you on this one?
Because, again, I am totally split on the benefits and the risks and the harms on
this one. I wasn't there a state getting ready to pass a law.

(19:35):
I want to say New York that was saying that there are three content areas that they
were going to prevent AI from answering prompts on.
And one of them was mental health.
One of them was medical.
And I believe legal.
I know legal just came up with one.
Yeah, I think there's a state trying to take action to prevent that mental health,

(19:57):
like I understand if you are in crisis and you need assistance, but but again.
You're not communicating with a human, you're communicating with a set of
instructions that all it's doing is trying to predict the next word.
It's not there is not cognitive ability behind it.

(20:19):
That that's where my issue lies with it.
Yeah, I mean, I used one of the AI tools this week for some health advice, and
it's something that like I wouldn't normally make a doctor's appointment, but
it's like a little gut check of, hey, should I should I be worried about, you
know, this headache or whatever it is?
And so it's hard to think about this one because we've we've been dealing with the

(20:43):
WebMD for many, many years.
But now these things are getting much more advanced and much more ingrained in our
culture. Should we be concerned about students turning to AI for medical and
health advice first before they talk to an adult?
I think adults are turning to AI for for mental health assistance and medical
assistance before going.

(21:04):
I mean, they've done it with what like you just said, Mark, for with WebMD for
years. I'm going to I'm going to say two things and I'm laughing internally because
I'm going to talk about my mental health, but I'm going to start it off with a
sponsor. So I guess I'm selling out my mental health.
But check out meter meter dot com slash K-12 tech talk.
Go to meter dot com slash K-12 tech talk to book a demo.

(21:25):
They can do your network infrastructure, your full stack network, Internet wired, Wi-Fi
cellular. They can just help you with your with all that you have going on.
They can optimize your operations all to scale.
So I agree.
Transition, by the way.
Yes. Phenomenal.
I'm trying.
I think, Josh, probably I'm feeling like we've unpacked this probably a couple of

(21:49):
times and Mark, maybe you and I have to.
I know it's I feel like maybe on the road to Indiana, Josh, I talked about my mental
stuff a little bit before I do this weird thing and I won't fully unpack it.
But I'll tell you that I've told some people, of course, I've told my wife I have
this thing in my head.
I do this like ritual thing where I start over.

(22:11):
I get these feelings where like when my day doesn't go right, when I when I'm told
something I don't like, my brain just wants to say like you failed.
OK. And I do this weird starting over thing.
I have particular things that I have to like check off a list to make myself feel
good. It's like this pursuit of perfectionism.
It's this weird, complicated thing.

(22:32):
OK, so my wife, the closest person to me, I've tried to unpack that with her over the
years. She doesn't get it.
OK, my parents don't get it.
Closest friends don't fully get it.
I spent and I don't know what that is.
And I could talk about it way more than I'm going to talk about it now.
I go to chat GPT.
I talk to chat GPT for a while saying all this junk that I'm saying, plus more.

(22:57):
It gave me some pretty good insight and resources to check out.
It gave me a little diagram on this little system that I get stuck in.
Wow. And it helped me.
It didn't fix me.
But that's like just just great example of like, yeah, because I is this thing that has
access to all the stuff, you know, not my wife, who is a nurse, but she doesn't have

(23:18):
access to all this mental health.
Who's this weird guy I married?
You know what's wrong with him?
Stuff. But GPT can help me.
So I'm completely with you on, of course, students, kids, we should do these guardrails.
But great for adults to just have access way better than a Google search and going down
like a subreddit and getting lost in it.

(23:40):
I'll have to find the video and send it to you there.
I came across a video, I think, on LinkedIn this week of it was an interview with Trevor
Noah and Simon Sinek, and they were talking about I can't remember the gist of it now,
but it was the difference between nurses who care and nurses who are like empathetic, I

(24:01):
think. And they were drawing the corollary with A.I.
that. The gist of it is A.I.
is not a human, you are not going to make a mental connection or a personal connection
with A.I., and you can draw that corollary with education, the whole point of teachers in

(24:24):
the room is to create that that human connection and interaction with the students in
their room. A.I.
cannot do that.
It will not do that.
Right, right.
So that is why the human in both of these cases, Chris, if you were to go to a
professional, a therapist or counselor, whatever, whatever that name is, they could

(24:46):
give you those same resources.
But you are making a human connection with that person.
You're not doing that with A.I.
But the concerns and the rightfully so concerns are that there are some students and
some adults that don't get that and they are having relationships, they are having
disconnections, they don't realize they're talking to a computer.

(25:07):
And that's a very, very risky, risky issue that we have to keep an eye on.
It's a one way connection.
It's not a it's yeah, it's not a two way connection.
Yeah, I believe when I when I was talking to chat GPT and I told my wife, like, hey,
she gets me and we had some, you know, talking about, I was speaking to chat as a

(25:28):
female. Why?
I was going to say, why is it a she?
Because she gets me. Oh, she understands my mental health.
Oh, my.
Oh, my. I think I'm a prime example.
Of what this is.
These dark thoughts.

(25:48):
All right, let's move along.
Yeah, this one went pretty far here.
All right, let's close it out with Clever's 2026 cyber security report.
We've covered this one every year for the last, I don't say three years or so.
And Clever comes out with an annual report on what are the major trends in cyber
security that districts are dealing with.
As you can imagine, cyber security is increasingly becoming a concern for

(26:11):
districts, but a very, very surprising number.
Fifty two percent of districts experience an incident in 2025.
That is up 31 percent in 2023.
That's a sharp, sharp increase in districts experiencing an incident.
That yeah, that's not cool.
The other number, which also saw an even greater increase, although I think we can

(26:33):
understand why. Last year, the number, the percentage of districts experiencing
vendor related incidents went from four percent in 2023 to 32 percent
last year. Now, we also did have one of the largest data breaches, vendor
breaches of power school. So I think that's a good portion of it.
But power school was obviously not the only one.

(26:53):
Some good news, though. We've talked about MFA.
In fact, the three of us just did a presentation on MFA.
Ninety seven percent of I.T.
staff and ninety three percent of teachers are using MFA.
So that's a fantastic, fantastic job.
Good job out there.
And the number we were presenting just last week on student MFA, we were showing the
number of five percent of students.
Well, now Clever is showing 13 percent of students are using MFA.

(27:18):
And those numbers are the same for elementary, middle and high school.
So great to see that those numbers are increasing.
And obviously, schools are concerned about cell phones and
not having cell phones to be able to implement MFA.
But our presentation, we all talked about the different cell phone list options that
are available, including Clever, ClassLink and RapidID.

(27:41):
Thirteen percent. That's pretty good.
What do you think the tipping point is?
What to be that push to get everybody on it?
Twenty five percent?
Thirty percent?
I think it's going to be your neighbor.
Once the district next door to you turns it on, then it's going to go.
I feel like I'm getting like close.

(28:02):
Well, I am.
Yeah. I don't want to go first, though, in the area.
Josh went first.
No, he didn't, though. He like chickened out.
Yeah, we haven't done it, done it.
It's yeah, I mean, you're committing to some things.
All right, well, that's it for the news this week.
All right. Well, a hot topic of the week, and I feel very timely because we got into

(28:26):
this earlier.
Do you guys outwardly block the NCAA tournament while you're streaming on your
networks?
Some context for this one, the three of us did a presentation on student account
security and passwords and authentication, and I feel like it was like Chris and I
and then Josh.
Yeah, polarized, block this, block this, say no to that, and then Chris and I was like,

(28:50):
well, I don't even remember what the conversation was.
It doesn't even matter.
It doesn't. It was everything.
It was you saying no to everything.
What was great, though, is your team was in the room to like one of my guys was in
the house. Yeah, yeah, yeah.
You can tell they drink the Josh Kool-Aid and they're like, oh, there's different
flavors of Kool-Aid.
Look at Chris and Mark.

(29:12):
There's a different world out there.
Wait, we don't have to say no.
The really funny thing is one of my former guys who now works at a different
district was in the room and they don't even do MFA for their staff.
So it was funny.
Yeah, I he would be great to just like ask questions like, hey, we know that you

(29:33):
work with Josh for years.
Yeah. What's it like not working for him anymore?
Like, well, no, I think I think the reality is that he's not working for me anymore.
Conversation. I think the real conversation is there because he went.
My district is relatively I won't say small.
It's bigger than Chris's district, but it's not crazy large.

(29:54):
And he went to a crazy large district.
I think the real conversation there from a comparative standpoint is from the
medium sized district to the uber large district.
What changes?
Like we were we are way more agile.
Like I can say we're turning something off in that afternoon.
It's getting turned off.
We're at that uber large district.
It's taking him months to turn something off.

(30:17):
Josh, what's the definition of crazy large district?
His district, I think, is like 15000 students.
That's one of the large top 10 in the state.
You were talking Missouri, Mark.
OK, OK. We're not nationalizing it.
Yeah, Mark.
But we were talking about that with because that that there is some greatness

(30:40):
in that a small, medium size, I would say up, up.
Yeah, yeah.
Well, in Missouri, when you start getting to the 10000 range, whatever,
that's when the red tape comes out.
Oh, yeah.
Even what I just said about, hey, I think I'm kind of close to doing clever MFA.
Yeah, it's just me deciding.
Right. Wait, waiting on Chris.
Right. When Chris is ready, he'll make the recommendation kind of thing.

(31:03):
It doesn't I mean, yes, it has to pass through a little bit there, but not a whole lot.
Yeah, I'm trying to remember what it was that we were talking about.
It might have been this email security tool.
I turned something on.
Oh, I think it might have been encryption email.
Like we just turned it on like DLP rules.
And you use the word secure in your subject line.

(31:25):
It's encrypting that message.
He was saying like that's got to go through layer after layer
after layer of conversation.
It's got to get all this approval.
They're filtering tool.
We use the same filtering tool they do.
And being able to turn on the screen monitoring feature
has taken years to get approval where we just turned it on.
Yeah. Hey, I like you're dodging the thing.

(31:48):
Do you allow streaming of this basketball thing going on?
Oh, yeah. I'm not in the office.
So if it works, I'm not there to turn it off.
What about your content filter?
Like have you have you explicitly blocked it?
I don't for staff.
I don't know.
I would have to look.
It probably works for staff on the content filtering right now.

(32:09):
Maybe not the FortiGate, but on Chromebooks at home, it probably works.
OK. Have you have you gone out of your way to block it?
We have not gone out of our way to block it.
Do you know how it works?
I, I believe it it works.
And we may or may not in our office have a bracket pulled up on a TV screen.

(32:30):
We we may or may not be doing like a tech maintenance competition.
Well, there was a bracket.
I started a bracket pool over on K-12 Tech Pro last week, too.
No money changing hands.
IRS, if you're listening.
No E-rate gifting going on.
No, no.
All right, Mark, you said you had some national federal laws,

(32:53):
I think, that were being discussed that would take the majority of the episode.
What you got?
Yeah, we have four major bills on student safety
and technology in the federal government right now.
Unsure if all four will go anywhere, but I wanted to go through all four.
I'm going to start from, I would say, the most bipartisan ready bill

(33:15):
and we'll go to the harder ones, the more complicated ones.
These aren't necessarily directly related to schools,
but there are there's a little bit of overlap
and it's related to a lot of the conversations we've had.
So I'll go through each bill one by one.
I'll tell you what the Republican
and Democrats are saying about this, and then I'll give you a chance to respond.
So I'm going to start with the easiest one.
And by the easiest one, I mean the one where both Republicans

(33:36):
and Democrats are pretty close to saying, yeah, we're thumbs up on this.
Sammy's Law, H.R.
2657, it's passed the House and the Energy and Commerce Committee,
and it's ready to go with some pretty close to bipartisan support.
This is named after a teenager who passed away
after buying fentanyl laced pills over Snapchat.

(33:57):
And so this is going to require large social media platforms
to allow third party child safety software to monitor accounts.
And if the software detects dangerous behavior like drug use or self-harm,
it has to notify parents.
So does not require age verification.
It assumes the platforms already know how old the users are.
Republicans are back in this bill.

(34:19):
Democrats are, for the most part, on board with this one.
So that is Sammy's Law very likely to go into into the next phase
of of bill pass with bipartisan support.
So I'll pause there, give you a chance to respond on and on on this bill.
I see no strong issue.
I see no issues with this.
Yeah, do it. Check.

(34:41):
Great. OK, well, that will send some bipartisan support on our team, too.
We'll let you decide who's who.
COPA 2.0.
So COPA is a 1998 privacy law protects students under the age of 13.
The gist of COPA 2.0 is we're moving that age from 13 to 16.

(35:03):
It's going to be banning platforms from serving any personalized ads to minors.
Creates a dedicated enforcement arm in the FTC focused on children's privacy
and gives families the right to request deletion, review
or correction of any sort of data.
So we're kind of just moving the age level from 13 to 16.
Republicans, by and large, support this one.

(35:24):
Democrats are not necessarily split, but they they are just asking
for a couple of revisions on this one.
I do think this one is going to be passing the Democrats and Republicans
just kind of need to get together in a room and hash out some of the details.
But you are seeing a little bit more bipartisan support on this one
than some of the other ones.
COPA 2.0.

(35:44):
I would say make it push it to 18.
Yeah, my issue, my issue with COPA
and the understanding of the populace of COPA
is this is a limitation placed on website providers, not school districts.
And I think school districts get put into the place of enforcement

(36:06):
here all the time, and they shouldn't be.
Yeah, this this is a law against the website provider, not the district.
So whatever that enforcement looks like, there needs to be clearer, clearer
statements around that to help people understand
that it is not the district's responsibility. Correct.

(36:29):
Love it. Give me 18. Give me 16.
Send it. Yeah, let's do it.
Of the four that we're talking about right now, while this one doesn't have
the same level of votes on both sides, it does have a clear path.
So I think the Democrats and Republicans, they really just need to get
a couple of lines hashed out.
And then we're going to see COPA 2.0 moving forward.
You say this bad boy's not been poked since 1998.

(36:52):
Mm hmm. Well, for what was for his birthday, Mark?
1970s, although there's been some revisions since then.
It's just turned 50.
It was 1976, so it just turned 50.
OK, poke that bad boy.
Let's get to the controversial ones.
The App Store Accountability Act, H.R.

(37:13):
three, one, four, nine.
This one targets Apple's App Store and Google Play, not the apps themselves.
It requires app stores to verify that a parent or guardian is at least 18 years
old before a minor can download an app.
Requires parental consent for each app that a student or child might download.
And it shifts the enforcement from Instagram,

(37:36):
TikTok, Facebook to the app stores themselves.
That's going to I'm kind of giving the the the the headline here away.
Republicans are excited about this one.
They want to move this one forward.
They want to stop kids from downloading apps in the first place.
Democrats, on the other hand, are saying, wait a minute.
We want the enforcement and the liability on the apps themselves.

(37:59):
What the Democrats are concerned about is this shifts the responsibility
upstream and gives TikTok, Instagram, Facebook a kind of a yeah. Yeah.
So they are opposed to this one.
And that's the main reason why is they want the tech companies
themselves to bear the responsibility, not the app stores.

(38:21):
I I could I could see that.
Mm hmm. Say that more.
I could land on had that responsibility of age verification
being with the app manufacturer, not the app store.
However, I could also get there on two layers,
the manufacturer and the app store.
Well, that's in and of itself is a controversial topic is age verification,

(38:45):
because in order to tell if a student or a child is under the age of 18 or 16
or 13, you need to know how old everybody is.
And so this is where we kind of get to a slippery slope of all.
That means that all of us adults have to demonstrate our age or prove our age.
Yeah. So, yeah, my kid's going to want to talk.

(39:06):
I got to fill out the stupid thing just to get it downloaded
and then want an account.
Yeah, I got to fill out more information.
Yeah. And upload your driver's license.
Get that trash out of here.
Trash this one.
The other concern is like you don't even have to have children
like these apps, these tools, the app stores, the the apps themselves.
This is just inconveniencing us.

(39:28):
Chris wants his TikToks and quickly.
I love how Chris goes to inconvenience.
I'm like, I don't want to upload my driver's license to some third party website.
But OK, that's a that's another one.
By the way, check out Incident IQ.
If you're leading an I.T.
department in your school district right now, you're listening to us.
You know, the job is not getting simpler.

(39:50):
You manage a bunch of devices.
You do a bunch of support tickets.
You're thinking about cybersecurity.
You're planning your refresh cycles, all that stuff.
And you're still supposed to be strategic.
That's where Incident IQ comes in.
When I.T. runs smoothly, your classrooms do, too.
So check out IncidentIQ.com on how they can help you.
Now, the last one is the biggest, but also has the most.

(40:14):
I would say divided Congress on this one.
So this is the Kids Act.
We've talked about various different bills that have made it through
Congress in different iterations.
They've really all been bundled up.
There's been 12 different child safety bills put into this one package
called the Kids Act.
The Kids Internet and Digital Safety Act is the official name.
This is divided on party lines completely.

(40:36):
So Republicans are in favor.
Democrats are against it.
This requires social media platforms to offer parental control tools,
bans platforms from targeting minors with ads for harmful or illegal products,
requires companies to submit annual compliance reports.
The FTC includes guardrails around AI chatbots and how they interact with kids
and does not require platforms to verify whether a user is actually a minor.

(41:00):
So why is this such a divided bill?
Why are Democrats against this one?
Republicans view this as the most comprehensive child
safety legislation ever brought to the House floor.
So they are very, very excited about this one.
They're very for it.
Democrats are saying, hang on a second.
You've actually removed liability from the tech companies.

(41:21):
You have removed duty of care.
You've put the burden on the parents.
Their biggest concerns around this one is that they have removed duty of care.
Essentially, a tech company just has to say, oh, we didn't know
our product was causing any sort of issues and they're no longer liable for any harms.
In addition to that, if a parent wanted to bring a lawsuit to that company,

(41:43):
they have to show that the company intentionally
tried to cause or knew that their product was causing harm.
How are you going to do that?
You need internal working documentation or emails.
That requires a whistleblower.
So essentially, you could have a company that knows their tech product
or their website is harming children.

(42:05):
But unless a parent has proof that that company knew about the harms,
there's really not much they can do to bring a lawsuit against the company.
And so Democrats on the left side are saying, hey, we're on board
with the spirit of this one, but you are taking the liability away
from the tech companies.
And that actually puts us in a more dangerous place than before.
So the Kids Act, the Kids Internet and Digital Safety Act.

(42:28):
Very, very split in Congress right now.
Unsure where this is going to go.
I think and this isn't just specific to the Kids Act.
I think any time you remove a responsibility of liability
from anyone, from any party, and you can go to some of the companies
that are refusing to sign data privacy agreements

(42:50):
without a limit of liability clause in the privacy agreement,
that's immediately a red flag to me.
There's there really should not be a discussion
about removing liability on a product like social media
or data privacy agreements or weapons, cars, whatever.

(43:11):
Like there's going to be a liability factor there.
And anytime you talk about removing a liability,
again, I think that's a red flag.
Yeah. Another concern they have, too, is that this kind of pulls away
states rights from establishing lines and barriers.
So New York and California have all passed laws,
really putting more of the onus in the in the burden on tech companies.

(43:34):
Well, this actually removes those state laws and puts a
in the Democrats' opinions, a lower bar for the tech companies
and puts parents at risk and kids at risk.
So I am unsure if the Kids Act is going to go anywhere.
Interestingly enough, the particular areas where they were concerned,

(43:54):
this had very good bipartisan support in the Senate.
They removed certain language that the Democrats are the most concerned about
when it got to the House.
So will it go back to the Senate version or will the House
try to push through this act that is very, very split on party lines?
Who knows?
Yeah, I feel like this is
these are things that shouldn't

(44:16):
make a room split like their thoughts and opinions.
Yeah. So the fact that they're split here, there's something wrong
because this is an easy vote.
Yes. Like if it were good, like student safety, child safety.
That's an easy win to come back and tell your constituents that,
hey, I did this thing.

(44:36):
So, yeah, there's if it's split, something needs to be looked at.
So I leave myself undecided on this one.
Yeah. Well, here's the thing.
This is like, you know, back in the days of smoking.
This is like everybody saying that we agree that cigarettes are bad for you.
The difference, though, is that the Democrats are saying
we want to make sure that we're holding the cigarette companies liable.

(44:59):
And that's the part of the bill that they're concerned about.
We all agree cigarettes are bad, and we all agree that there's
some really bad stuff on the Internet and some social media
products that are not good for kids.
How we enforce it.
That's where we're seeing the division. Interesting.
Mm hmm. Mm hmm.
All right. Well, guys, there's another round of games

(45:20):
getting ready to get started, and we've gone a little long.
You are. So, Josh, you have your camera off.
Yeah, and I'm going to I can turn it back on.
No, I'm going to put this on the record, though.
I believe when your camera is off, you speak more. Maybe
because you want to be seen.
You know, your camera's off, so you just want to be seen. OK.

(45:41):
Josh is making me want to take off tomorrow.
I'm really thinking about it.
It's beautiful weather. It's going to be gorgeous tomorrow.
We're going to be on the boat.
Can I come? Yeah, come on down.
Mark, you want to come down? I'll be right there.
Hey, speaking of us getting back together.
So three sponsors to get through Fortinet email Fortinet podcast at Fortinet dot com.

(46:02):
They do your FortiGate, your FortiToken, your FortiAnalyzer
and all their other Forti products.
They were at Midwest Tech Talk.
Great sponsors there, but great sponsors of the podcast, too.
So email Fortinet podcast at Fortinet dot com.
We saw Dominic Mayer there as well.
A proud sponsor of the podcast.
You can email Dominic with Extreme Networks.

(46:22):
D Mayer at Extreme Networks dot com.
Get your extreme switching on.
And last but not least is Lightspeed.
They can do your content filter.
They talked about a cool insights
module that they have that you can know what apps and websites,
different things you're your people are using.
Their signal product is kicker insight tool is awesome.

(46:44):
I my guy Matt did that session and he came back to me wanting us to do insights.
And all that said, though, boys,
we're going to be hanging out in Chicago, thanks to our sponsors
and our listeners at Kosen April 13th through the 15th.
We're going to have the table set up.
Should be a great time.
And I'm looking forward to hanging out with you guys.

(47:05):
We I mean, it was pretty fast paced hanging out Midwest Tech Talk.
Yeah, you were running around like a crazy person.
But yeah, we Mark and I hung out.
Mark and I did did.
Here's a here's a burning question I have as we wrap up.
Mark and I hugged in the middle of our session.
Yeah. As he left.
Did you hug? Did you hug?

(47:27):
Yeah, I got hugged in front of the we hugged in front of an audience.
Yeah. You know what I thought was interesting about that, too?
No one asked why we hugged and why Mark disappeared.
No, they didn't either.
It was just so weird.
It's like it makes it those weren't weird moments.
Yeah. But to an outside person, we were doing a presentation
and I had done the math that I needed to leave halfway through a presentation.

(47:50):
So while we were presenting, I randomly hugged Chris and Josh
and then bolted out of the room, which there was no side door.
I had to walk straight through the middle of everybody.
I believe you like you had to walk past the projection.
Yeah, it was clearly seen.
Yeah. Mark is leaving.
There's a person clapping as I walked out the door.
It was great. Just hug those guys and left abruptly.

(48:14):
Anyways, I'm looking forward to hanging out with Kosen
because we will have some downtime being at the table together.
Maybe we'll have we'll have to go to what the Italian beef place again.
Yes. Mr. Beef. Mr. Beef.
We love beef.
That's the end.
We might not be the same.

(48:35):
You share the same pain that I do.
The views and opinions expressed on the K-12 Tech Talk podcast
are the personal opinions of Josh, Chris and Mark, and do not represent
the views or opinions of our sponsors or other organizations that we're affiliated with.
The material and information presented here is for general information only.
We do not sell or advertise any products or services.

(48:56):
The material and information presented here is for general information
and entertainment purposes only.
Thanks for listening and we'll see you next week.

All Episodes

Episode 256 - AI in the Classroom and the Federal Bills Shaping Student Safety

Episode Transcript

Popular Podcasts

Stuff You Should Know

Dateline NBC

Kingdom of Fraud

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}Episode 256 - AI in the Classroom and the Federal Bills Shaping Student Safety