AI Avalanche: Test Your Luck in the Tundra
Explore the transformative potential and challenges of Artificial Intelligence in education through an interactive game-show format. Engage in an avalanche of real-time scenarios demonstrating AI's impact on personalization and differentiation, weighing its risks, and decide: Is integrating AI into teaching testing your luck, or a strategic step towards a smarter classroom?
Alright. Thank you, everybody. Appreciate you joining us. My name is Ryan Lufkin, vice president of global academic strategy strategy here at Instructure. I'm excited about this session. I think it's gonna be one of the more engaging and interactive sessions of the entire conference, and my job is literally just to kick it off and get off the stage.
So, thank you for joining us. Please, be active participants. And with that, I'm gonna hand it off to doctor Book. Thank you so much, Brian. Good afternoon, everyone.
Welcome to beautiful Las Vegas. I have a beautiful game ahead of us. Welcome to AI avalanche test your luck in the tundra. Awesome. I love it.
Where there's an applause button somewhere. I love it. Alright. So That's what I was waiting for. During this game show, we are gonna brave the icy challenges of integrating AI in education.
I'm your host, doctor Rachel Book, guiding you through today's game show panel. Let's meet our contestants, Charlie, Dave, and Nate before we dive into the game. Charlie. Hi, everyone. I'm Charlie Thayer, chief academic officer at Lincoln Learning Solutions.
Born and raised as a social studies teacher and have worked in education for about a decade and a half, in a number of different capacities and super excited to be on this game show and definitely win the game. Challenge accepted. Hello, everyone. My name's Dave Whitehead. I'm the chief technology officer at Lincoln Learning Solutions.
I've been at Lincoln Learning for fourteen years, been in IT for twenty three years, and have about fifteen years in management, and still growing in leadership experience. It's an honor and a pleasure to speak with you all today. Hey, everybody. I'm Nate Ober. I am an AI and ML strategist at AWS.
Before working for AWS, I was a chief technology and product officer for a number of education, companies, ed tech companies. And before that, I grew up at the Utah State Office of Education as a, education specialist working in formative assessment. Really glad to be here tonight. Thanks so much for having me. Thank you, contestants, for those great introductions.
It's fantastic to have such experienced professionals with us here this evening. Before we get started, just a little bit of housekeeping. If you have received one of these cards, you'll want to make sure that you get it stamped before you leave. If you have not, you'll want to get one of these because they're really cool prizes associated. Second housekeeping, when you hear the Yeti summoned, please watch out for flying objects.
And the third housekeeping that we wanna cover tonight is we want your engagement as much as we want the contestants on the stage. So there are gonna be some questions, poll questions throughout. Please scan the QR code so you're prepared to answer those questions. Alright, contestants. It's looking like we have a pretty excited crowd here today.
I came to see a Yeti. That's awesome. Alright. I think we're ready to start test your luck. Charlie, you're gonna start us off.
Personalization in education. Charlie, here's your question. How can AI be tailored to support the diverse learning needs and styles of students without creating a one size fits all approach? I love this question. The the possibilities, with AI, generative AI to be more specific, are really, endless, if you have the right mindset. Being able to revolutionize the way that we create content and create content more quickly, is something that we've never had capacity to do, the way that we can with generative AI.
The big thing I think that is really important with tailoring content for each individual learner is obviously understanding where that learner is, where their strengths are, where the weaknesses are, what kinds of things, what kinds of supports, that student, needs, at the end of the day. And being able to, you know, leverage artificial intelligence to be able to great, better understand that, where that student is, how they learn, what they do like, what they don't like, is a really critical aspect of that, and then paired with, being able to understand those things, being able to then adapt content but, not just letting the computer run loose, and and spawn new content for us, that would be very easy for us to do, but would be very, very, very, I think, irresponsible of us as well. Let me say that overseeing a team of folks, that would greatly benefit from the, ability to tailor content both in the instructional capacity and in the curriculum development capacity. We see it, to be used as a tool, and just that. Humans should remain in control, of the tool, and be the final decision maker at the end of the day because they're not they're gonna know that student better than anyone else.
So, I think we have more possibilities when it comes to to generative AI or artificial intelligence more broadly to understand learners, and then adapt content to meet those needs of those students so that we're not just homogenizing education. I I hear all the time I've heard a lot lately, rather, I should say, not a one size fits all, one size fits one. So being able to really hone in on each individual student, and the capabilities that AI allow us, are I'm super excited for, what the future holds, given that we take a very responsible approach to it moving forward. Well done, Charlie. Thanks for starting us off on such a good note.
Let's see if Dave or Nate have anything to add. Yeah. I'll just add to that real quick. I think, when we can infuse the power of AI technology within platforms, historical and real time data can help tailor a personalized experience for the student or the educator. While existing machine learning and legacy algorithms may have worked for personalization, the advancements of AI will increase the speed, accuracy, and flexibility for unique personalization needs? I've been working, in formative assessment, or I worked in formative assessment for a really, really long time, and there are some problems that we could never solve, or at least not add adequately.
So, imagine if your interest is in, say, sports or your interest is in computer technology and the the the content that you're receiving is talking about farming and you don't know anything about farming. It's gonna be a lot harder for you to find places in your scaffolding to hang that information than if you can customize the information to your to the to the learner's interests. And these are things that I worked on even way back in form of assessment when I was working at the Utah State Office of Education on the UTIP system was how do you create content that fits the learner the individual learners, interests to make it more engaging and interesting. And I think now we have the opportunity to understand better the learner's interests, their capabilities, to look at the scaffolding that we already have with the sort of the learning standards, and find the the gaps in one particular learner's, understanding and then create custom tailored content for them in a way that was never possible before. So for me, I am, geekily, very nerdily excited for what generative AI can do to help with, student learning.
Great responses, everyone. Charlie, you do get the point for this question. Dave, your turn. AI ethics and privacy. Dave, how can educational institutions ensure their IT systems are scalable and adaptable to emerging AI technologies? I think to give a few key points, educational institutions can ensure scalability and adaptability by investing in cloud based solutions such as Amazon Web Services.
They can also foster, partnerships with technology providers and continuously train IT staff on the latest IT advancements. The Yeti has been summoned. I I think I'd just like to add a little bit of flavor of context around that too. So, when I, was working at Questar assessment, on high stakes, high stakes assessment, we I visited schools across the country and one of the challenges that I saw that we faced even back then was sort of wireless or network connectivity for all the students who wanted to get involved in all the technology that they needed to. And and more and more, all the technology, all of the content that we have is available on the cloud, through, cloud delivered solutions, often through, provided through AWS.
In order to access that content, learners need to have fast, computers connected to fast and, Internet connections. And so I think that's the the first thing. One other thing that I think is super important is security. So our customers choose, AWS because we're the most, secure platform, or the most the most, secure, cloud, provider that's available today. And, I think when you're considering all the uses for AI and especially, like, the ethics behind AI, security should be one of your first considerations.
Dave, it was a solid, solid answer, but Nate has stolen your point. I think AWS I think AWS knows something about scalability, and adaptability. So, of course Audience, it's your turn. Let's weigh in. Go ahead and start your poll.
Which feature is most crucial for AI to effectively support diverse learning styles? Give you just a couple of minute seconds. Alright. Great responses. Keep them coming in if you're still taking a minute. Panel, do we have anything to add to the, poll responses that the audience has shared with us? Adaptive learning algorithms is at fifty eight, user centric design is twenty seven and I can't read it.
Multimodal Content, fourteen percent. For me, this one this this one's kind of a tough one, because I think in their own ways, each one of these is, pretty critical, to diverse learning styles. I think you can make arguments for any one of these honestly, to support those learners in the different ways that they learn. So I yeah. I I can agree and disagree with every, every option that's up here.
Yeah. I kinda weigh in on the adaptive learning, system and algorithms one. You guys obviously feel the same way. I think it includes a combination of intelligent chatbot tutors potentially, adaptive testing, personalized learning pass. You can adjust the, content delivery based on learners goals and current knowledge state as well.
So, systems like this, they can collect data on learners interactions, performance, and preferences, and, AI algorithms can then analyze this data to understand each learner's unique style and progress. Yeah. I I think that's right. We've been talking already about this a lot just tonight and being able to create personalized content for learners, it's clear that everyone here believes that too. But based on where the learner's at in their education progression, their knowledge acquisition, being able to remediate on the fly based on some of these algorithms is super important.
And as we all know, being able to remediate quickly is and the power of AI is exactly that to be able to remediate, quickly. So I agree. Good call, audience. Great insights from both the audience and our panel. Thank you, everyone.
Nate, it's your turn to hit the button. Curriculum and AI integration. Nate, what ethical consideration should be taken into account when implementing AI technologies that collect and analyze student data? I really think about this in sort of three key categories. So first is bias, second is equity of access, and third is security. So, for for bias, I think we all know we have to be cautious about what comes out of, a generative content in particular.
It's colloquially known as hallucination, but it's real. It's it's common for a machine learning language or or, for generative AI to make up answers and be very confident and defend those, those answers vigorously. And so I think we have to be really careful because some of those answers are gonna be biased. Like, the the challenge another challenge with generative learning is that it was trained by on a dataset which is the Internet, which is, extremely biased in it in its own right. So, if you look at some of the Reddit channels out there, there's there's a strong degree of bias.
If you look at Twitter, don't look anymore because it's called x, but if you looked at x, and, you you, examined the content there, you'd find a lot of sort bias because humans are are generally biased and therefore, when you're training a large language model on the content of the Internet, you're gonna uncover, you're gonna find some some bias. And so I think we have to be very sensitive to that. I think for equity of access, generative AI, like we don't not all of us think about it like this, but I feel very lucky to have this sort of upbringing, this sort of, I've had computers since I was a a little kid. Right? Like, I have a lot of access to technology that not everybody has access to. And in many of our like, it's hard for me to believe because I live and breathe this stuff, but there's I talk to people every day who have never even heard of, say, ChatCPT.
I shouldn't say that out loud, but, there are a lot of people out there who don't know what generative, technology is, and and I think we have to be cautious because if you don't if you don't watch out for equity, there's gonna be a greater divide between the people who have the power of say a calculator and people who don't, and that's a huge huge difference. And then I think as far as security goes, you may not be consciously aware of it, but you're sharing a lot of yourself when you ask even the most basic question or you have the basic the most basic conversation large language model. And so I think we have to be very cautious about where that data goes, what it's gonna be used for, and and, and, and how it's gonna safe how your safe your privacy is gonna be safeguarded so that that information isn't somehow used nefariously or against you. So those are, for me, the the big things that I that I think about related to to this. Nate's response has given us a glimpse into the icy cavern of AI, but Dave has summoned the Yeti to add a little more light on the topic.
Dave, can you help illuminate the dark corners of this question? Yeah. Very well articulated, Nate. Appreciate you. I guess we know the ethical use of student data is critical when incorporating AI and, clear policies should be established that protect student privacy and ensure data security. Even consent from parents and students should also probably be considered and be essential before collecting any data.
Also, transparency about how AI, uses this data is necessary to maintain trust. We should only collect data that serves educational purposes and ensure it's handled with the utmost confidentiality, so to speak, a little bit to security. Regular audits and compliance with data protection laws should be routine, to safeguard our students' information. As AI continues to emerge, as a pivotal tool in enhancing student engagement and driving educational outcomes, we must also prioritize the privacy and security of our students data. Imagine a classroom where where every student receives a personalized learning experience tailored to their unique needs and learning pace.
AI can make this possible by analyzing those patterns. To be able to balance the the personalization with user and data privacy, I think aligning frameworks like COPPA and FERPA will be critical as well. These regulations provide a critical foundation for protecting student information, ensuring From encrypted databases to secure access, every step in the From encrypted databases to secure access, every step in the AI process should be designed with the utmost security in mind. Protecting user data should be non negotiable. I think in essence, AI in education is not just about pressing a button and watching the magic happen.
It's a thoughtful integration of technology, pedagogy, and ethical standards. It's about enhancing the educational journey, while holding the privacy and security of our educators and students at the highest priority. Dave, very insightful. You stole your point back. I'm on the board.
It's time for the audience to answer another poll. When implementing AI in education, what should be prioritized to ensure it's both equitable and ethical? Panel, if you'll take a look when the responses start coming in, we'd like to hear your input as well. Looks like there's a battle between comprehensive access and privacy strategy and transparent data policies at the moment. Gentlemen, any thoughts? From a comprehensive access and privacy strategy, I think it's essential to ensure, equitable and ethical use. Strategy should balance the need for broad access to educational resources with robust data privacy and security measures.
Implementation of strict data privacy policies that comply with regulations, like I mentioned earlier, and providing informed consent for data collection to ensure transparency on how data is used. Also, the usage of techniques to anonymize student data so that personal information can't be traced back to individual students. I I agree with this one. I think this kind of speaks to equity, at the end of the day. And I think, even beyond AI, I'll step back to the pandemic.
I think that showed us that, there really is a lack of equity in online learning, maybe learning in general, across the board. There really just is not a level playing field yet. Hopefully, we can get there one day, but I think, we see the same we see that same case when it comes to artificial intelligence as well, varying policies, various strategies, it continually evolving as we're trying to keep up with it and different stances on whether or not, we want to adopt it or we don't to adopt it. We're afraid of it. We're leaning into it.
It's it's just kind of a a very big cross section right now of, AI and where it's at, its evolution, and I think, I I think that one's spot on. Yeah. I I think the audience knows best here. I mean, it seems like largely you all agree with what we've been saying, which is I think being thoughtful about the way that we implement our access policies to make sure that everyone has, equitable access. I think about privacy.
I mean, I I take a lot of personal, I put a lot of thought and work into my personal privacy and I do the same thing as I'm thinking about my customers and about especially learners and have my whole life set. It seems like, the audience agrees with us and I think that shows they're well informed. Wonderful. Charlie, it is your last chance to press your luck. Teaching with AI.
Charlie, in what ways can AI enhance the role of teachers and how can we ensure it doesn't replace the human element that is crucial for learning? So this might sound, this this might not jive, it might jive. I I think Be careful, Charlie. There's teachers in the audience. I know. This is not against teachers.
This is actually one hundred percent in favor of teachers and humans in general. I think artificial intelligence will be hard pressed to fully replace the composite that a human can bring to education or any other relationship for that matter. I think as much as it evolves and as much as it continues to get better and better and better until we get to the point of something really crazy like actually client cloning humans, I think we're still gonna be a a few degrees separated. And I think that's probably a good thing. I think, if, if anything, I believe core to the values of my initial training as a teacher and the work that I did, as a teacher in the classroom and then moving into online learning is, that human connection is critical to a student success.
So many different facets of education and learning tie back to, that human connection being there. And it can be subtle. It can be very, very subtle, but it can be almost a breaking point sometimes in a measure of success, when a student is struggling or when a student is excelling or when a student is just moving through their course. I think I think AI will be hard pressed to to to replace that. I'm not answering the question though.
So after I'm off my soapbox, I do think that there are a number of ways, that AI can help chip away at the growing laundry list of things that teachers have to, take care of, on a daily basis. Even if it's as simple as providing grading suggestions or being able to can you go back to some of the other questions that we were talking about adapting and enhancing content? One of things that Nate was talking about a few minutes ago made me think, you know, the baseball example. If I have a classroom of thirty different kiddos, that all have specific interests, I can't do that many more times than one at a time. And great teachers will do it. They'll try to do it.
They're they're they're they will do it. I think AI gives us a broader sword to wield, as educators in the classroom to really meet those students where they are and really engage their interests. So I think, just taking away some of the the administrative or clerical tasks that a teacher has to to conduct on a day to day basis, AI should be able to help with that. But again, I do not believe that it is a replacement for teachers. I do not, believe that it should be a replacement for teachers ever, honestly.
I I and I'm a I'm a very big believer in what AI can do for us, and how it can help accelerate us. But again, it is a tool. It is just another tool, very powerful tool. With great power comes great responsibility to use the the Spider Man reference. Mhmm.
We have to use it responsibly. We have to use it, ethically. It can do a lot of bad as much as it can do good. So I think finding creative ways and having an open mind to continue to learn, continue to learn, continue continue to understand because I think if if we shy away from it just because we're afraid of it and it's different, we will get kind of left behind from it. And that's even more dangerous I think than being acclimated with the powers that AI brings to the table.
So I think that comes in the way of very strong and continual professional development. I think the past twelve to eighteen months, in the, advancements of AI and some of the capabilities, that it possesses today that it didn't in the months before, where we are right now, it's moving quick. It's moving real real quick. So it's incumbent upon us I think to remain aware, of what what it is and what it can do, and looking for ways to, help us do our jobs faster, help us do our jobs quicker, more efficiently. All of those things, I think of the phrase work smarter not harder if there's something that we don't have to beat our heads against the wall as teachers and AI can help take some of that out of our way, great.
That's a that's a phenomenal use of it, but again, not a replacement for a human. There are some things that maybe humans don't have to do. It is the twenty first century. We can we have conveniences. This is another convenience.
Well, Charlie, you hit that snowball out of the park. Nate or Dave, do you have anything to add? I think Charlie nailed it. He said it perfectly. AI is just a tool. It's no different, conceptually than your smartphone or even a calculator.
So human monitoring and involvement is key on this one. I think I'm just gonna double down and and say, two things. Undifferentiated heavy lifting is the first thing, and I can't remember the second thing, so I'll come back to that one. So at AWS, we have this theory about removing the undifferentiated heavy lifting from companies. And what that means is there's a lot of stuff that companies do in their day to day that doesn't add any value to their bottom to their customers to their end customers' needs, right, or to the end users' needs.
So what is running a server and doing patches, on that server do for your end user? Nothing. It's transparent to them. They have no idea. I think the same thing is true for AI and education and teachers. Right? Teachers every day do a lot of undifferentiated heavy lifting.
They don't they don't spend all of their time focused on helping that one individual learner get the help that they need in order to be successful, in order to grasp that one topic that they were having a really hard time with. Instead, they have to go through and, you know, do, well, you know, grade a lot of papers or write a lot of content, instead of just having something that can ask the teacher what what he or she wants to build and then have that content custom tailored to the to the teacher's interest as well as the learner's best interests. So I think that's, that's the first thing. Oh, and I remember the second thing. Human in the loop is another thing that you're gonna hear often when it comes to developing generative AI systems.
And for me, what that means, and it's important to do that for most things at this point, we can't just deliver a large language model to a learner and expect good outcomes. The content that comes out of it will be biased as we talked about. The content that comes out of it may hallucinate as we talked about. And so having a filter on top of that generative model like a teacher, somebody who knows what the point of the instruction is and what it is that the what that the student or the learner should be getting out of it. I think that's super important.
I don't think that, I think the last thing, and I could go down a rabbit hole on this, but we as humans learn from mirroring. We mirror the people that are the, that, are the people that we want to be when we grow up. We imitate them. We are imitators. That's how we learn.
You can't imitate and learn from a large language model. There's a lot of, communication that happens that isn't verbal. There's a lot of communication that happens that isn't written and you can't get those things from a large language model. Thank you all for those wonderful answers. Charlie, you did receive the point for this question.
Dave, it's your turn to press your luck. I happen to know this is Dave's favorite topic, cybersecurity in education. Thanks, Rachel. It is, it is a topic I'm very fond of. Pretty much any organization, especially ones using AI technologies, maintaining robust cybersecurity is paramount.
It's not just about safeguarding data. It's about protecting the trust and safety of our students and educators. So to cover a few things, regular cybersecurity awareness training is essential. It empowers every individual in any organization with the knowledge to recognize and prevent potential cyber threats. Training should, also cover the latest trends in cyber attacks and the best practices for digital hygiene.
It's about creating a culture of cyber security mindfulness. Data protection and least privilege access is also a key factor. By adhering to strict data protection protocols, you can ensure that sensitive information is not only accessible to authorized users, it's protected against unauthorized ones, from any sort of breaches. Implementing least privilege access is another critical practice. It means that users are granted only access or rights they need to perform in their roles, nothing more.
This minimizes the risk of insider threats and reduces the potential of damage, from any security incidents. Predictive analytics, another facet of AI, can help anticipate cybersecurity incidents before they happen. By processing vast amounts of data and recognizing the signs of potential attack, AI enables us to put safeguards in place, ensuring that our students and educators data remain secure. Another significant advancement, in AI's ability to automate responses, to security incidents in the fast paced world of cyber threats. Timing is very, very critical.
AI driven systems can respond to threats in real time often without even the need for human intervention. It's not only speeds up response times, but also allows IT staff to focus on strategic initiatives rather than constantly firefighting. AWS can help with all these technical aspects to manage the ongoing threat landscape. We know from experience. I feel as we embrace the benefits of AI in education, we must continue to fortify our defenses.
By implementing these best practices, we create a secure foundation for educational organizations to thrive in the digital era. If you're interested in hearing more about EdTech and cybersecurity, come see my two PM session tomorrow, SecureEd, Cybersecurity and AI for Safer EdTech Environments. Awesome answer, Dave. Nate, Charlie, anything to add? I think he, I think he did a really great job. The, I think the only thing that I'd double down on is AWS is really great at this.
So does everybody know who Amazon Web Services is AWS by show of hands yet? Alright. Good. So, AWS, as many of you may know, is architected to be the most secure cloud computing environment available today. And that doesn't mean that we'll cover all of the challenges. Companies like Lincoln Learning need to do their very you know, to build their software, to architect their systems to be secure as well.
But it's a solid foundation from which to build, and so I think you said it very well. I think AWS also has, some really great services that help you analyze your current security footprint and posture, to improve and improve very quickly with simple clicks of a button. You can go from, whatever score and really, really improve your overall posture. It's fantastic. Good.
And I think the only thing that I would add to this, from someone who doesn't sit maybe in, in an, as a technical role as these gentlemen, is, really from the educator standpoint, knowing some of the inherent risks that are involved in using AI tools when it comes to student data, student privacy, and things like that. Knowing the right questions to ask, or just things to be aware of of when you're starting to use different tools or whether maybe your students are starting to use different tools, through the classroom as well. Because depending on the tool, when you put things out there, they are out there. And there are there are risks, there there are definitely risks to that. So maybe not knowing the depth, depth and breadth, that these guys know, when it comes to cybersecurity, but knowing, indeed that there are risks to be aware of, and to be, to remain mindful of so that, you don't just slap a solution in place.
And I know we're not doing that, but I was excited to go out and start playing with AI when when it came out as well. And I ran off and created me an account and started plugging things in and showed my wife and she was not impressed, at all. She said this is creepy. Get it away from me. And I said, oh, alright.
But just knowing and being aware, that we have to be mindful of those kinds of things, I think, is is very critical as as educator from an educator standpoint, not necessarily from a technologist standpoint, that we have to be we have to be aware of those things. Thanks for the additional input. Dave, you received the point for that question. Alright. Our third and final poll question for the audience.
What aspect of cybersecurity do you believe needs the most focus to protect AI systems in educational settings? Oops. Alright. End user training and data encryption security protocols are battling back and forth here as our audience participation decreases. Any any comments, we've one more question to go guys. I know we're rolling on time and wanna ask questions.
So anything to add for this one? No, I agree. Yeah. Same. You guys you guys are now in it. I think I kind of just said as much.
Yes. Awesome. So we are aligned which I like very much. Nate, you are up for the final question. Future of jobs and AI education.
What are the long term implicate implications of AI on the job market and how should education systems adapt to prepare students for these changes? Oh, good. I got the easy one. So, I'm the father of three. I am a licensed educator in Utah, and I've worked in education companies my whole life. And, I wish I could tell you exactly how to think about this.
But what I can tell you is that it's not going away, that, generative AI tool sets are here. And your students, or if you're, principal, your instructors, your teachers, or if you're an admin, your entire staff. They're all using these tools whether or not you endorse them. And I think what will make a learner more successful in life is if they know how to use them effectively and can diagnose fact from fiction and do their own, research to determine what is truth and what is not truth. That's the first thing.
I think the the second thing I would say is being super, being a good human and knowing about other humans and being able to work well with other people, and being able like knowing where you come from, knowing, who you're gonna work with, we're always forever going to have to work with each other. That's not gonna go away either. And so I think teaching learners to have a lifelong learning goal, like continue to learn about new technologies, continue to learn about new things, continue to to, grow and and to become smarter, to be a good person, have good emotional intelligence, be able to spend time, with other people and be able to communicate well, I think those things are gonna serve you well, And, be familiar with all the tools that are gonna be available to you because they will help you progress as a person in your career as you, continue. Thanks, Nate for challenging that icy question. Dave, Charlie, do you dare add anything to his answer? It's hard to say.
I agree. I it's it's it's a dicey question. The main thing I think is it is indeed, I agree, it's not going anywhere. It is going to change the way that we work though. Not to say that it necessarily flux and change in recent years.
So, I think it's going to be a long, long, long play, but I do indeed believe that it's going to change the way that we work, in the near term and certainly in the long term. Yeah. I agree with my colleagues on this one. Not not too much to add. Thank you, gentlemen.
Nate, you received the point for that question. Well, folks, we have a tied game and it brings us to the end of our game show. I wanna thank our contestants and our audience for participating in AI avalanche. It's been a thrilling journey through the icy terrains of AI and education. If you would like to receive more information, you can scan this QR code.
We've taken all of the insights from our panelists and some research and put them into a document that we would like to share out with all of you. So there's gonna be a form on there to fill out. You can also visit us, we're all at booth eighty one if you have questions directly for, Charlie or Dave. And I do think we have a couple of minutes for questions. Yep.
We have about five minutes for questions if anybody has any questions. One last there. Zach, will you run this? I'm sorry. I can't hear you. So I'm gonna thank you so much.
Sorry. Give us one second. Sorry. Sorry. Go ahead.
Where were you at? Oh. I almost forgot. Okay. We're ready we're ready for the question. This is dangerous.
They gave me a microphone. So, first of all, welcome to Nevada. Glad you're here, home means Nevada. Secondly, I was just curious if you could say anything about the use of AI and potentially with instructional design in that combination? I think you've alluded to teaching with the tools, but then what if you have another, like, formal system applied using AI, Gen AI or otherwise? Any thoughts on that from the panel? I mean our contestants? I do. Only if it gets me an extra point and I can steal the game at the end.
I can be bribed. Yes. Yes. Kinda along the same lines as what we've talked about, we have dabbled, and we're we we have a couple projects in motion right now where we've leveraged gen AI to help support just kinda content development, not necessarily as much instructional design, though I do think that there is a place for that as well. Once you have something that you can train those models on to understand what your instructional design strategies and practices are.
But we've just kind of dipped our toe in the water. We've dabbled with a few things. We are a curriculum development, org. That that's what we do. But one of the big things that we wanted to be wary of from the very beginning is that we don't just turn into a a coffee grinder where we're churning out curriculum and, feeding it out of the AI machine.
So we've been very discretionary, about our approaches, very deliberate with our approaches, and really try to be, mindful of running things through that review process. We were at a conference a few weeks ago and they were talking about, some of the practices that another organization is putting in place. And I I if I if it wouldn't have been inappropriate, I would have stood up and cheered in the middle of it, because I one hundred percent agreed with what they were talking about, and it was much of what we've said here today. Keeping the human in the loop. I love that.
Right? Like that, keeping them involved in the practices just to aid. I think we can get to the point where we can leverage it through instructional design or with instructional design. I think it's gonna take some time for us to get there, to to leverage it accordingly. And I think it would be the only responsible way for us to do it. Bit bit bit bit bite sized pieces.
Thank you, Charlie. Any other questions? Did I get the quarter point? Maybe. Please throw a Yeti in. Yeah. Alright.
We do have one more question. While we're getting the microphone over for this question, if you have a card that needs punched, you can see Zach or take a picture of this screen and go to our booth. We'll punch it there as well so you don't have to wait here. Either way yes, sir. Thank you.
Hi. Thank you. Jeff Simmons from the universe University of San Diego. I'm an instructional designer. I thought that question regarding enhancing, possibly replacing teachers at some point was an intriguing one.
And I kinda wish there was a survey for that, but, in the meantime, isn't it a more nuanced question? I mean, when we're looking around here, we're seeing some of the most engaged teachers available. But in America, that's not always the case. Where wherever you live, if we look at the average education levels in certain areas, and I know in in our minds, many of us are thinking k through twelve, but I work in an adult, online situation. So I think it may depend on who we're focusing on. If we're looking at, for example, take auto driven cars and Las Vegas.
The average driver here, depending on their state at the time, maybe maybe it's best to have auto driven at certain places. And and so in a similar sense, I'm wondering whether or not AI, having only been out for a year and a half exponentially increasing. If they reach AGI in the end of this decade or soon after, we're already matching human intelligence and personal assistance and all of these things. Plus, kids, depending on their generation, make friends with people they don't even know. And in my generation, it was always face to face.
So I guess that was really broad coming at you, but I'm wondering what your take is because you didn't say never, but it kind of felt that way. And I felt in certain situations, it would not only enhance, but in online cases, maybe replace online faculty sooner than later? Thank you. Wonderful question. I think Nate is chomping at the bit to answer this. I'm really excited to talk about this one actually.
I've thought a lot about this. So I worked at a couple of places that, provide asynchronous learning for, learners who otherwise wouldn't get access for any number of reasons. Either they can't make a schedule work. I worked at a company called StraighterLine, which provided low cost college rated classes, that were fully transferable through ACE. And those were all asynchronous.
There was no instructor involvement at all. What would make them more valuable to the learner is to have a virtual tutor that could interact with them, that they could ask questions of, that they could do more of a Socratic learning style, of education with. Do I think that that would replace a teacher? I don't think that that will replace a large portion of what teachers do for the learner, having engaged with CHAT GPT or with, Claude, which is my preferred model. Like, I I don't see that as replacing a human interaction anytime in the near future. I'm not saying never, but I don't see it anytime in the near future.
In another place where I, where I think it's valuable is in I also worked for a company called Origin, formerly APDS, which served content, educational access to incarcerated learners. And there, the the the number of teachers to student, it's improbable that a student will get any access. There's a wait list that is multi years in order to get access to many, many of these, classes. And so if you can augment some of these asynchronous options that are currently available today with a large language model or some other kind of virtual tutor, I think it will create more equitable access to those who are the neediest among us, but I don't think that it will replace what you think of as our current education system anytime in the near future. Mostly because, as I mentioned, I think we we do a lot of sort of mirroring and a lot of that, we we comprehend we we retain information better when we, can relate to the person who's instructing us.
I have a couple pieces. One, I really appreciate the question, because I do think it's, gives us a lot to think about. One of the things that I think, concerning the question though, I think it comes down to a matter of equity and what our definition or what our measure of equity is, in these different instances. So is an AM model, is it truly, truly, I think we owe it to ourselves to ask what is that measure of equity, at the end of the day. And and do we feel that it's truly equitable? Or or are we even equipped to be self aware enough to know that it is equitable, at the end of the day.
So I think that's something that we probably have to consistently look back to to see what is that measure of equity, because I think that's a more complex equation than a plus b equals c, at the end of the day. Thank you both for that answer. So I know that dinner waits on the other side of this, this, session. So I do wanna thank you again for your time, and we would love to speak with you, further. Please come visit us so that we can continue this conversation.
Thank you all. Thanks everyone. Enjoy the rest of your conference.
So, thank you for joining us. Please, be active participants. And with that, I'm gonna hand it off to doctor Book. Thank you so much, Brian. Good afternoon, everyone.
Welcome to beautiful Las Vegas. I have a beautiful game ahead of us. Welcome to AI avalanche test your luck in the tundra. Awesome. I love it.
Where there's an applause button somewhere. I love it. Alright. So That's what I was waiting for. During this game show, we are gonna brave the icy challenges of integrating AI in education.
I'm your host, doctor Rachel Book, guiding you through today's game show panel. Let's meet our contestants, Charlie, Dave, and Nate before we dive into the game. Charlie. Hi, everyone. I'm Charlie Thayer, chief academic officer at Lincoln Learning Solutions.
Born and raised as a social studies teacher and have worked in education for about a decade and a half, in a number of different capacities and super excited to be on this game show and definitely win the game. Challenge accepted. Hello, everyone. My name's Dave Whitehead. I'm the chief technology officer at Lincoln Learning Solutions.
I've been at Lincoln Learning for fourteen years, been in IT for twenty three years, and have about fifteen years in management, and still growing in leadership experience. It's an honor and a pleasure to speak with you all today. Hey, everybody. I'm Nate Ober. I am an AI and ML strategist at AWS.
Before working for AWS, I was a chief technology and product officer for a number of education, companies, ed tech companies. And before that, I grew up at the Utah State Office of Education as a, education specialist working in formative assessment. Really glad to be here tonight. Thanks so much for having me. Thank you, contestants, for those great introductions.
It's fantastic to have such experienced professionals with us here this evening. Before we get started, just a little bit of housekeeping. If you have received one of these cards, you'll want to make sure that you get it stamped before you leave. If you have not, you'll want to get one of these because they're really cool prizes associated. Second housekeeping, when you hear the Yeti summoned, please watch out for flying objects.
And the third housekeeping that we wanna cover tonight is we want your engagement as much as we want the contestants on the stage. So there are gonna be some questions, poll questions throughout. Please scan the QR code so you're prepared to answer those questions. Alright, contestants. It's looking like we have a pretty excited crowd here today.
I came to see a Yeti. That's awesome. Alright. I think we're ready to start test your luck. Charlie, you're gonna start us off.
Personalization in education. Charlie, here's your question. How can AI be tailored to support the diverse learning needs and styles of students without creating a one size fits all approach? I love this question. The the possibilities, with AI, generative AI to be more specific, are really, endless, if you have the right mindset. Being able to revolutionize the way that we create content and create content more quickly, is something that we've never had capacity to do, the way that we can with generative AI.
The big thing I think that is really important with tailoring content for each individual learner is obviously understanding where that learner is, where their strengths are, where the weaknesses are, what kinds of things, what kinds of supports, that student, needs, at the end of the day. And being able to, you know, leverage artificial intelligence to be able to great, better understand that, where that student is, how they learn, what they do like, what they don't like, is a really critical aspect of that, and then paired with, being able to understand those things, being able to then adapt content but, not just letting the computer run loose, and and spawn new content for us, that would be very easy for us to do, but would be very, very, very, I think, irresponsible of us as well. Let me say that overseeing a team of folks, that would greatly benefit from the, ability to tailor content both in the instructional capacity and in the curriculum development capacity. We see it, to be used as a tool, and just that. Humans should remain in control, of the tool, and be the final decision maker at the end of the day because they're not they're gonna know that student better than anyone else.
So, I think we have more possibilities when it comes to to generative AI or artificial intelligence more broadly to understand learners, and then adapt content to meet those needs of those students so that we're not just homogenizing education. I I hear all the time I've heard a lot lately, rather, I should say, not a one size fits all, one size fits one. So being able to really hone in on each individual student, and the capabilities that AI allow us, are I'm super excited for, what the future holds, given that we take a very responsible approach to it moving forward. Well done, Charlie. Thanks for starting us off on such a good note.
Let's see if Dave or Nate have anything to add. Yeah. I'll just add to that real quick. I think, when we can infuse the power of AI technology within platforms, historical and real time data can help tailor a personalized experience for the student or the educator. While existing machine learning and legacy algorithms may have worked for personalization, the advancements of AI will increase the speed, accuracy, and flexibility for unique personalization needs? I've been working, in formative assessment, or I worked in formative assessment for a really, really long time, and there are some problems that we could never solve, or at least not add adequately.
So, imagine if your interest is in, say, sports or your interest is in computer technology and the the the content that you're receiving is talking about farming and you don't know anything about farming. It's gonna be a lot harder for you to find places in your scaffolding to hang that information than if you can customize the information to your to the to the learner's interests. And these are things that I worked on even way back in form of assessment when I was working at the Utah State Office of Education on the UTIP system was how do you create content that fits the learner the individual learners, interests to make it more engaging and interesting. And I think now we have the opportunity to understand better the learner's interests, their capabilities, to look at the scaffolding that we already have with the sort of the learning standards, and find the the gaps in one particular learner's, understanding and then create custom tailored content for them in a way that was never possible before. So for me, I am, geekily, very nerdily excited for what generative AI can do to help with, student learning.
Great responses, everyone. Charlie, you do get the point for this question. Dave, your turn. AI ethics and privacy. Dave, how can educational institutions ensure their IT systems are scalable and adaptable to emerging AI technologies? I think to give a few key points, educational institutions can ensure scalability and adaptability by investing in cloud based solutions such as Amazon Web Services.
They can also foster, partnerships with technology providers and continuously train IT staff on the latest IT advancements. The Yeti has been summoned. I I think I'd just like to add a little bit of flavor of context around that too. So, when I, was working at Questar assessment, on high stakes, high stakes assessment, we I visited schools across the country and one of the challenges that I saw that we faced even back then was sort of wireless or network connectivity for all the students who wanted to get involved in all the technology that they needed to. And and more and more, all the technology, all of the content that we have is available on the cloud, through, cloud delivered solutions, often through, provided through AWS.
In order to access that content, learners need to have fast, computers connected to fast and, Internet connections. And so I think that's the the first thing. One other thing that I think is super important is security. So our customers choose, AWS because we're the most, secure platform, or the most the most, secure, cloud, provider that's available today. And, I think when you're considering all the uses for AI and especially, like, the ethics behind AI, security should be one of your first considerations.
Dave, it was a solid, solid answer, but Nate has stolen your point. I think AWS I think AWS knows something about scalability, and adaptability. So, of course Audience, it's your turn. Let's weigh in. Go ahead and start your poll.
Which feature is most crucial for AI to effectively support diverse learning styles? Give you just a couple of minute seconds. Alright. Great responses. Keep them coming in if you're still taking a minute. Panel, do we have anything to add to the, poll responses that the audience has shared with us? Adaptive learning algorithms is at fifty eight, user centric design is twenty seven and I can't read it.
Multimodal Content, fourteen percent. For me, this one this this one's kind of a tough one, because I think in their own ways, each one of these is, pretty critical, to diverse learning styles. I think you can make arguments for any one of these honestly, to support those learners in the different ways that they learn. So I yeah. I I can agree and disagree with every, every option that's up here.
Yeah. I kinda weigh in on the adaptive learning, system and algorithms one. You guys obviously feel the same way. I think it includes a combination of intelligent chatbot tutors potentially, adaptive testing, personalized learning pass. You can adjust the, content delivery based on learners goals and current knowledge state as well.
So, systems like this, they can collect data on learners interactions, performance, and preferences, and, AI algorithms can then analyze this data to understand each learner's unique style and progress. Yeah. I I think that's right. We've been talking already about this a lot just tonight and being able to create personalized content for learners, it's clear that everyone here believes that too. But based on where the learner's at in their education progression, their knowledge acquisition, being able to remediate on the fly based on some of these algorithms is super important.
And as we all know, being able to remediate quickly is and the power of AI is exactly that to be able to remediate, quickly. So I agree. Good call, audience. Great insights from both the audience and our panel. Thank you, everyone.
Nate, it's your turn to hit the button. Curriculum and AI integration. Nate, what ethical consideration should be taken into account when implementing AI technologies that collect and analyze student data? I really think about this in sort of three key categories. So first is bias, second is equity of access, and third is security. So, for for bias, I think we all know we have to be cautious about what comes out of, a generative content in particular.
It's colloquially known as hallucination, but it's real. It's it's common for a machine learning language or or, for generative AI to make up answers and be very confident and defend those, those answers vigorously. And so I think we have to be really careful because some of those answers are gonna be biased. Like, the the challenge another challenge with generative learning is that it was trained by on a dataset which is the Internet, which is, extremely biased in it in its own right. So, if you look at some of the Reddit channels out there, there's there's a strong degree of bias.
If you look at Twitter, don't look anymore because it's called x, but if you looked at x, and, you you, examined the content there, you'd find a lot of sort bias because humans are are generally biased and therefore, when you're training a large language model on the content of the Internet, you're gonna uncover, you're gonna find some some bias. And so I think we have to be very sensitive to that. I think for equity of access, generative AI, like we don't not all of us think about it like this, but I feel very lucky to have this sort of upbringing, this sort of, I've had computers since I was a a little kid. Right? Like, I have a lot of access to technology that not everybody has access to. And in many of our like, it's hard for me to believe because I live and breathe this stuff, but there's I talk to people every day who have never even heard of, say, ChatCPT.
I shouldn't say that out loud, but, there are a lot of people out there who don't know what generative, technology is, and and I think we have to be cautious because if you don't if you don't watch out for equity, there's gonna be a greater divide between the people who have the power of say a calculator and people who don't, and that's a huge huge difference. And then I think as far as security goes, you may not be consciously aware of it, but you're sharing a lot of yourself when you ask even the most basic question or you have the basic the most basic conversation large language model. And so I think we have to be very cautious about where that data goes, what it's gonna be used for, and and, and, and how it's gonna safe how your safe your privacy is gonna be safeguarded so that that information isn't somehow used nefariously or against you. So those are, for me, the the big things that I that I think about related to to this. Nate's response has given us a glimpse into the icy cavern of AI, but Dave has summoned the Yeti to add a little more light on the topic.
Dave, can you help illuminate the dark corners of this question? Yeah. Very well articulated, Nate. Appreciate you. I guess we know the ethical use of student data is critical when incorporating AI and, clear policies should be established that protect student privacy and ensure data security. Even consent from parents and students should also probably be considered and be essential before collecting any data.
Also, transparency about how AI, uses this data is necessary to maintain trust. We should only collect data that serves educational purposes and ensure it's handled with the utmost confidentiality, so to speak, a little bit to security. Regular audits and compliance with data protection laws should be routine, to safeguard our students' information. As AI continues to emerge, as a pivotal tool in enhancing student engagement and driving educational outcomes, we must also prioritize the privacy and security of our students data. Imagine a classroom where where every student receives a personalized learning experience tailored to their unique needs and learning pace.
AI can make this possible by analyzing those patterns. To be able to balance the the personalization with user and data privacy, I think aligning frameworks like COPPA and FERPA will be critical as well. These regulations provide a critical foundation for protecting student information, ensuring From encrypted databases to secure access, every step in the From encrypted databases to secure access, every step in the AI process should be designed with the utmost security in mind. Protecting user data should be non negotiable. I think in essence, AI in education is not just about pressing a button and watching the magic happen.
It's a thoughtful integration of technology, pedagogy, and ethical standards. It's about enhancing the educational journey, while holding the privacy and security of our educators and students at the highest priority. Dave, very insightful. You stole your point back. I'm on the board.
It's time for the audience to answer another poll. When implementing AI in education, what should be prioritized to ensure it's both equitable and ethical? Panel, if you'll take a look when the responses start coming in, we'd like to hear your input as well. Looks like there's a battle between comprehensive access and privacy strategy and transparent data policies at the moment. Gentlemen, any thoughts? From a comprehensive access and privacy strategy, I think it's essential to ensure, equitable and ethical use. Strategy should balance the need for broad access to educational resources with robust data privacy and security measures.
Implementation of strict data privacy policies that comply with regulations, like I mentioned earlier, and providing informed consent for data collection to ensure transparency on how data is used. Also, the usage of techniques to anonymize student data so that personal information can't be traced back to individual students. I I agree with this one. I think this kind of speaks to equity, at the end of the day. And I think, even beyond AI, I'll step back to the pandemic.
I think that showed us that, there really is a lack of equity in online learning, maybe learning in general, across the board. There really just is not a level playing field yet. Hopefully, we can get there one day, but I think, we see the same we see that same case when it comes to artificial intelligence as well, varying policies, various strategies, it continually evolving as we're trying to keep up with it and different stances on whether or not, we want to adopt it or we don't to adopt it. We're afraid of it. We're leaning into it.
It's it's just kind of a a very big cross section right now of, AI and where it's at, its evolution, and I think, I I think that one's spot on. Yeah. I I think the audience knows best here. I mean, it seems like largely you all agree with what we've been saying, which is I think being thoughtful about the way that we implement our access policies to make sure that everyone has, equitable access. I think about privacy.
I mean, I I take a lot of personal, I put a lot of thought and work into my personal privacy and I do the same thing as I'm thinking about my customers and about especially learners and have my whole life set. It seems like, the audience agrees with us and I think that shows they're well informed. Wonderful. Charlie, it is your last chance to press your luck. Teaching with AI.
Charlie, in what ways can AI enhance the role of teachers and how can we ensure it doesn't replace the human element that is crucial for learning? So this might sound, this this might not jive, it might jive. I I think Be careful, Charlie. There's teachers in the audience. I know. This is not against teachers.
This is actually one hundred percent in favor of teachers and humans in general. I think artificial intelligence will be hard pressed to fully replace the composite that a human can bring to education or any other relationship for that matter. I think as much as it evolves and as much as it continues to get better and better and better until we get to the point of something really crazy like actually client cloning humans, I think we're still gonna be a a few degrees separated. And I think that's probably a good thing. I think, if, if anything, I believe core to the values of my initial training as a teacher and the work that I did, as a teacher in the classroom and then moving into online learning is, that human connection is critical to a student success.
So many different facets of education and learning tie back to, that human connection being there. And it can be subtle. It can be very, very subtle, but it can be almost a breaking point sometimes in a measure of success, when a student is struggling or when a student is excelling or when a student is just moving through their course. I think I think AI will be hard pressed to to to replace that. I'm not answering the question though.
So after I'm off my soapbox, I do think that there are a number of ways, that AI can help chip away at the growing laundry list of things that teachers have to, take care of, on a daily basis. Even if it's as simple as providing grading suggestions or being able to can you go back to some of the other questions that we were talking about adapting and enhancing content? One of things that Nate was talking about a few minutes ago made me think, you know, the baseball example. If I have a classroom of thirty different kiddos, that all have specific interests, I can't do that many more times than one at a time. And great teachers will do it. They'll try to do it.
They're they're they're they will do it. I think AI gives us a broader sword to wield, as educators in the classroom to really meet those students where they are and really engage their interests. So I think, just taking away some of the the administrative or clerical tasks that a teacher has to to conduct on a day to day basis, AI should be able to help with that. But again, I do not believe that it is a replacement for teachers. I do not, believe that it should be a replacement for teachers ever, honestly.
I I and I'm a I'm a very big believer in what AI can do for us, and how it can help accelerate us. But again, it is a tool. It is just another tool, very powerful tool. With great power comes great responsibility to use the the Spider Man reference. Mhmm.
We have to use it responsibly. We have to use it, ethically. It can do a lot of bad as much as it can do good. So I think finding creative ways and having an open mind to continue to learn, continue to learn, continue continue to understand because I think if if we shy away from it just because we're afraid of it and it's different, we will get kind of left behind from it. And that's even more dangerous I think than being acclimated with the powers that AI brings to the table.
So I think that comes in the way of very strong and continual professional development. I think the past twelve to eighteen months, in the, advancements of AI and some of the capabilities, that it possesses today that it didn't in the months before, where we are right now, it's moving quick. It's moving real real quick. So it's incumbent upon us I think to remain aware, of what what it is and what it can do, and looking for ways to, help us do our jobs faster, help us do our jobs quicker, more efficiently. All of those things, I think of the phrase work smarter not harder if there's something that we don't have to beat our heads against the wall as teachers and AI can help take some of that out of our way, great.
That's a that's a phenomenal use of it, but again, not a replacement for a human. There are some things that maybe humans don't have to do. It is the twenty first century. We can we have conveniences. This is another convenience.
Well, Charlie, you hit that snowball out of the park. Nate or Dave, do you have anything to add? I think Charlie nailed it. He said it perfectly. AI is just a tool. It's no different, conceptually than your smartphone or even a calculator.
So human monitoring and involvement is key on this one. I think I'm just gonna double down and and say, two things. Undifferentiated heavy lifting is the first thing, and I can't remember the second thing, so I'll come back to that one. So at AWS, we have this theory about removing the undifferentiated heavy lifting from companies. And what that means is there's a lot of stuff that companies do in their day to day that doesn't add any value to their bottom to their customers to their end customers' needs, right, or to the end users' needs.
So what is running a server and doing patches, on that server do for your end user? Nothing. It's transparent to them. They have no idea. I think the same thing is true for AI and education and teachers. Right? Teachers every day do a lot of undifferentiated heavy lifting.
They don't they don't spend all of their time focused on helping that one individual learner get the help that they need in order to be successful, in order to grasp that one topic that they were having a really hard time with. Instead, they have to go through and, you know, do, well, you know, grade a lot of papers or write a lot of content, instead of just having something that can ask the teacher what what he or she wants to build and then have that content custom tailored to the to the teacher's interest as well as the learner's best interests. So I think that's, that's the first thing. Oh, and I remember the second thing. Human in the loop is another thing that you're gonna hear often when it comes to developing generative AI systems.
And for me, what that means, and it's important to do that for most things at this point, we can't just deliver a large language model to a learner and expect good outcomes. The content that comes out of it will be biased as we talked about. The content that comes out of it may hallucinate as we talked about. And so having a filter on top of that generative model like a teacher, somebody who knows what the point of the instruction is and what it is that the what that the student or the learner should be getting out of it. I think that's super important.
I don't think that, I think the last thing, and I could go down a rabbit hole on this, but we as humans learn from mirroring. We mirror the people that are the, that, are the people that we want to be when we grow up. We imitate them. We are imitators. That's how we learn.
You can't imitate and learn from a large language model. There's a lot of, communication that happens that isn't verbal. There's a lot of communication that happens that isn't written and you can't get those things from a large language model. Thank you all for those wonderful answers. Charlie, you did receive the point for this question.
Dave, it's your turn to press your luck. I happen to know this is Dave's favorite topic, cybersecurity in education. Thanks, Rachel. It is, it is a topic I'm very fond of. Pretty much any organization, especially ones using AI technologies, maintaining robust cybersecurity is paramount.
It's not just about safeguarding data. It's about protecting the trust and safety of our students and educators. So to cover a few things, regular cybersecurity awareness training is essential. It empowers every individual in any organization with the knowledge to recognize and prevent potential cyber threats. Training should, also cover the latest trends in cyber attacks and the best practices for digital hygiene.
It's about creating a culture of cyber security mindfulness. Data protection and least privilege access is also a key factor. By adhering to strict data protection protocols, you can ensure that sensitive information is not only accessible to authorized users, it's protected against unauthorized ones, from any sort of breaches. Implementing least privilege access is another critical practice. It means that users are granted only access or rights they need to perform in their roles, nothing more.
This minimizes the risk of insider threats and reduces the potential of damage, from any security incidents. Predictive analytics, another facet of AI, can help anticipate cybersecurity incidents before they happen. By processing vast amounts of data and recognizing the signs of potential attack, AI enables us to put safeguards in place, ensuring that our students and educators data remain secure. Another significant advancement, in AI's ability to automate responses, to security incidents in the fast paced world of cyber threats. Timing is very, very critical.
AI driven systems can respond to threats in real time often without even the need for human intervention. It's not only speeds up response times, but also allows IT staff to focus on strategic initiatives rather than constantly firefighting. AWS can help with all these technical aspects to manage the ongoing threat landscape. We know from experience. I feel as we embrace the benefits of AI in education, we must continue to fortify our defenses.
By implementing these best practices, we create a secure foundation for educational organizations to thrive in the digital era. If you're interested in hearing more about EdTech and cybersecurity, come see my two PM session tomorrow, SecureEd, Cybersecurity and AI for Safer EdTech Environments. Awesome answer, Dave. Nate, Charlie, anything to add? I think he, I think he did a really great job. The, I think the only thing that I'd double down on is AWS is really great at this.
So does everybody know who Amazon Web Services is AWS by show of hands yet? Alright. Good. So, AWS, as many of you may know, is architected to be the most secure cloud computing environment available today. And that doesn't mean that we'll cover all of the challenges. Companies like Lincoln Learning need to do their very you know, to build their software, to architect their systems to be secure as well.
But it's a solid foundation from which to build, and so I think you said it very well. I think AWS also has, some really great services that help you analyze your current security footprint and posture, to improve and improve very quickly with simple clicks of a button. You can go from, whatever score and really, really improve your overall posture. It's fantastic. Good.
And I think the only thing that I would add to this, from someone who doesn't sit maybe in, in an, as a technical role as these gentlemen, is, really from the educator standpoint, knowing some of the inherent risks that are involved in using AI tools when it comes to student data, student privacy, and things like that. Knowing the right questions to ask, or just things to be aware of of when you're starting to use different tools or whether maybe your students are starting to use different tools, through the classroom as well. Because depending on the tool, when you put things out there, they are out there. And there are there are risks, there there are definitely risks to that. So maybe not knowing the depth, depth and breadth, that these guys know, when it comes to cybersecurity, but knowing, indeed that there are risks to be aware of, and to be, to remain mindful of so that, you don't just slap a solution in place.
And I know we're not doing that, but I was excited to go out and start playing with AI when when it came out as well. And I ran off and created me an account and started plugging things in and showed my wife and she was not impressed, at all. She said this is creepy. Get it away from me. And I said, oh, alright.
But just knowing and being aware, that we have to be mindful of those kinds of things, I think, is is very critical as as educator from an educator standpoint, not necessarily from a technologist standpoint, that we have to be we have to be aware of those things. Thanks for the additional input. Dave, you received the point for that question. Alright. Our third and final poll question for the audience.
What aspect of cybersecurity do you believe needs the most focus to protect AI systems in educational settings? Oops. Alright. End user training and data encryption security protocols are battling back and forth here as our audience participation decreases. Any any comments, we've one more question to go guys. I know we're rolling on time and wanna ask questions.
So anything to add for this one? No, I agree. Yeah. Same. You guys you guys are now in it. I think I kind of just said as much.
Yes. Awesome. So we are aligned which I like very much. Nate, you are up for the final question. Future of jobs and AI education.
What are the long term implicate implications of AI on the job market and how should education systems adapt to prepare students for these changes? Oh, good. I got the easy one. So, I'm the father of three. I am a licensed educator in Utah, and I've worked in education companies my whole life. And, I wish I could tell you exactly how to think about this.
But what I can tell you is that it's not going away, that, generative AI tool sets are here. And your students, or if you're, principal, your instructors, your teachers, or if you're an admin, your entire staff. They're all using these tools whether or not you endorse them. And I think what will make a learner more successful in life is if they know how to use them effectively and can diagnose fact from fiction and do their own, research to determine what is truth and what is not truth. That's the first thing.
I think the the second thing I would say is being super, being a good human and knowing about other humans and being able to work well with other people, and being able like knowing where you come from, knowing, who you're gonna work with, we're always forever going to have to work with each other. That's not gonna go away either. And so I think teaching learners to have a lifelong learning goal, like continue to learn about new technologies, continue to learn about new things, continue to to, grow and and to become smarter, to be a good person, have good emotional intelligence, be able to spend time, with other people and be able to communicate well, I think those things are gonna serve you well, And, be familiar with all the tools that are gonna be available to you because they will help you progress as a person in your career as you, continue. Thanks, Nate for challenging that icy question. Dave, Charlie, do you dare add anything to his answer? It's hard to say.
I agree. I it's it's it's a dicey question. The main thing I think is it is indeed, I agree, it's not going anywhere. It is going to change the way that we work though. Not to say that it necessarily flux and change in recent years.
So, I think it's going to be a long, long, long play, but I do indeed believe that it's going to change the way that we work, in the near term and certainly in the long term. Yeah. I agree with my colleagues on this one. Not not too much to add. Thank you, gentlemen.
Nate, you received the point for that question. Well, folks, we have a tied game and it brings us to the end of our game show. I wanna thank our contestants and our audience for participating in AI avalanche. It's been a thrilling journey through the icy terrains of AI and education. If you would like to receive more information, you can scan this QR code.
We've taken all of the insights from our panelists and some research and put them into a document that we would like to share out with all of you. So there's gonna be a form on there to fill out. You can also visit us, we're all at booth eighty one if you have questions directly for, Charlie or Dave. And I do think we have a couple of minutes for questions. Yep.
We have about five minutes for questions if anybody has any questions. One last there. Zach, will you run this? I'm sorry. I can't hear you. So I'm gonna thank you so much.
Sorry. Give us one second. Sorry. Sorry. Go ahead.
Where were you at? Oh. I almost forgot. Okay. We're ready we're ready for the question. This is dangerous.
They gave me a microphone. So, first of all, welcome to Nevada. Glad you're here, home means Nevada. Secondly, I was just curious if you could say anything about the use of AI and potentially with instructional design in that combination? I think you've alluded to teaching with the tools, but then what if you have another, like, formal system applied using AI, Gen AI or otherwise? Any thoughts on that from the panel? I mean our contestants? I do. Only if it gets me an extra point and I can steal the game at the end.
I can be bribed. Yes. Yes. Kinda along the same lines as what we've talked about, we have dabbled, and we're we we have a couple projects in motion right now where we've leveraged gen AI to help support just kinda content development, not necessarily as much instructional design, though I do think that there is a place for that as well. Once you have something that you can train those models on to understand what your instructional design strategies and practices are.
But we've just kind of dipped our toe in the water. We've dabbled with a few things. We are a curriculum development, org. That that's what we do. But one of the big things that we wanted to be wary of from the very beginning is that we don't just turn into a a coffee grinder where we're churning out curriculum and, feeding it out of the AI machine.
So we've been very discretionary, about our approaches, very deliberate with our approaches, and really try to be, mindful of running things through that review process. We were at a conference a few weeks ago and they were talking about, some of the practices that another organization is putting in place. And I I if I if it wouldn't have been inappropriate, I would have stood up and cheered in the middle of it, because I one hundred percent agreed with what they were talking about, and it was much of what we've said here today. Keeping the human in the loop. I love that.
Right? Like that, keeping them involved in the practices just to aid. I think we can get to the point where we can leverage it through instructional design or with instructional design. I think it's gonna take some time for us to get there, to to leverage it accordingly. And I think it would be the only responsible way for us to do it. Bit bit bit bit bite sized pieces.
Thank you, Charlie. Any other questions? Did I get the quarter point? Maybe. Please throw a Yeti in. Yeah. Alright.
We do have one more question. While we're getting the microphone over for this question, if you have a card that needs punched, you can see Zach or take a picture of this screen and go to our booth. We'll punch it there as well so you don't have to wait here. Either way yes, sir. Thank you.
Hi. Thank you. Jeff Simmons from the universe University of San Diego. I'm an instructional designer. I thought that question regarding enhancing, possibly replacing teachers at some point was an intriguing one.
And I kinda wish there was a survey for that, but, in the meantime, isn't it a more nuanced question? I mean, when we're looking around here, we're seeing some of the most engaged teachers available. But in America, that's not always the case. Where wherever you live, if we look at the average education levels in certain areas, and I know in in our minds, many of us are thinking k through twelve, but I work in an adult, online situation. So I think it may depend on who we're focusing on. If we're looking at, for example, take auto driven cars and Las Vegas.
The average driver here, depending on their state at the time, maybe maybe it's best to have auto driven at certain places. And and so in a similar sense, I'm wondering whether or not AI, having only been out for a year and a half exponentially increasing. If they reach AGI in the end of this decade or soon after, we're already matching human intelligence and personal assistance and all of these things. Plus, kids, depending on their generation, make friends with people they don't even know. And in my generation, it was always face to face.
So I guess that was really broad coming at you, but I'm wondering what your take is because you didn't say never, but it kind of felt that way. And I felt in certain situations, it would not only enhance, but in online cases, maybe replace online faculty sooner than later? Thank you. Wonderful question. I think Nate is chomping at the bit to answer this. I'm really excited to talk about this one actually.
I've thought a lot about this. So I worked at a couple of places that, provide asynchronous learning for, learners who otherwise wouldn't get access for any number of reasons. Either they can't make a schedule work. I worked at a company called StraighterLine, which provided low cost college rated classes, that were fully transferable through ACE. And those were all asynchronous.
There was no instructor involvement at all. What would make them more valuable to the learner is to have a virtual tutor that could interact with them, that they could ask questions of, that they could do more of a Socratic learning style, of education with. Do I think that that would replace a teacher? I don't think that that will replace a large portion of what teachers do for the learner, having engaged with CHAT GPT or with, Claude, which is my preferred model. Like, I I don't see that as replacing a human interaction anytime in the near future. I'm not saying never, but I don't see it anytime in the near future.
In another place where I, where I think it's valuable is in I also worked for a company called Origin, formerly APDS, which served content, educational access to incarcerated learners. And there, the the the number of teachers to student, it's improbable that a student will get any access. There's a wait list that is multi years in order to get access to many, many of these, classes. And so if you can augment some of these asynchronous options that are currently available today with a large language model or some other kind of virtual tutor, I think it will create more equitable access to those who are the neediest among us, but I don't think that it will replace what you think of as our current education system anytime in the near future. Mostly because, as I mentioned, I think we we do a lot of sort of mirroring and a lot of that, we we comprehend we we retain information better when we, can relate to the person who's instructing us.
I have a couple pieces. One, I really appreciate the question, because I do think it's, gives us a lot to think about. One of the things that I think, concerning the question though, I think it comes down to a matter of equity and what our definition or what our measure of equity is, in these different instances. So is an AM model, is it truly, truly, I think we owe it to ourselves to ask what is that measure of equity, at the end of the day. And and do we feel that it's truly equitable? Or or are we even equipped to be self aware enough to know that it is equitable, at the end of the day.
So I think that's something that we probably have to consistently look back to to see what is that measure of equity, because I think that's a more complex equation than a plus b equals c, at the end of the day. Thank you both for that answer. So I know that dinner waits on the other side of this, this, session. So I do wanna thank you again for your time, and we would love to speak with you, further. Please come visit us so that we can continue this conversation.
Thank you all. Thanks everyone. Enjoy the rest of your conference.
Related Content
On-Demand Webinars
-
On-Demand Webinars
-
Product Demos & Videos