AI Literacy for Science Teachers With Guest Amanda Bickerstaff of AI for Education [Episode 201]

Rebecca Joyner [00:00:01]:
If you have been listening to the Secondary Science Amplified podcast over the last few years, but especially this past year specifically, then you will know that I’ve been on an AI discovery journey, if you will. And this is because I started seeing a ton of questions from you all about AI. And I knew then, like, I had to become more educated about it, like I had to talk about it. And so through this process, this journey for me, I’ve been interviewing teachers using AI, as you’ve seen on the podcast, and I’ve been exploring the tools myself. And in my research about AI, I came across Amanda Bickerstaff of AI for Education and y’, all, the second I saw her website and saw that she was a former high school science teacher, I said we have to have her on the Secondary Science Simplified podcast. I knew we would all benefit from hearing from her. And so I’m so thrilled to say today’s the day she’s on the podcast and I think you’re going to just learn so much from her and her experience. So first let me introduce you to her.

Rebecca Joyner [00:01:08]:
Amanda is the co founder and CEO of AI for Education. She’s a former high school biology teacher and an EdTech executive with over 20 years of experience in the education sector. She has a deep understanding of the challenges and opportunities that AI can offer. She’s a frequent consultant, speaker and writer on the topic of AI in education, leading workshops and professional learning across both K12 and higher education. Amanda is committed to helping educators, staff and students maximize their potential through the ethical and equitable adoption of AI. In this episode, y’, all, Amanda busts some major misconceptions teachers, myself very much included, have around AI. She really shares about the importance of AI literacy, how to use it effectively, and also the anticipated long term impact on AI on the field of education. And with that, best practices for both science teachers and students alike.

Rebecca Joyner [00:02:12]:
Her background as a high school biology teacher is honestly, it’s truly the icing on the cake for me. And it was just so fun to dialogue with someone who knows way more about AI and technology than I literally ever could, but at the same time who also, I feel like, gets me and thus will get us in our little corner of the Internet here with it’s not Rocket Science and Secondary Science Simplified. Like, I feel like she gets us as secondary science teachers on a fundamental level because of her past career teaching science. So I just know I learned so much from this conversation and I really, really hope you do too. So without further ado, here’s my conversation with Amanda Bickerstaff of AI for Education. This is Secondary Science Simplified, a podcast for secondary science teachers who want to engage their students and simplify their lives. I’m Rebecca Joyner from It’s Not Rocket Science. As a high school science teacher turned curriculum writer, I am passionate about helping other science teachers, love their jobs, serve their students, and do it all in only 40 hours a week.

Rebecca Joyner [00:03:18]:
Are you ready to rock the time you spend in your classroom and actually have a life outside of it? You are in the right place, teacher friend. Let’s get to today’s episode. Hi, Amanda, how are you?

Amanda Bickerstaff [00:03:34]:
I’m good. Nice meet you. To, to, to be with you on a very gloomy day here in New York City, but excited to have this conversation.

Rebecca Joyner [00:03:41]:
I’m so excited to talk to you. I actually found you literally just through Google search. I was on my, like, own journey to try to learn more about using AI as a secondary science educator, and you came up and so I’m so excited to let my audience learn from you and all about your AI for Education program.

Amanda Bickerstaff [00:04:01]:
Well, I’m glad that Google is doing some good and glad that we get to do this as I’m a former science teacher.

Rebecca Joyner [00:04:07]:
Yeah.

Amanda Bickerstaff [00:04:07]:
And so that’s definitely a big area that I still think a lot about.

Rebecca Joyner [00:04:11]:
Well, and that’s. I was like, so tickled when I got to your website and looked at the team and I read your bio and you said you were. I was like, okay, we have got to have her. So I’m just thrilled you’re here. And I have to know, what got you into this space? Like, how did you transition from science teaching into now this AI space?

Amanda Bickerstaff [00:04:28]:
Well, I mean, it’s been a long journey. I mean, I taught right out of college and so I taught in the, in the Bronx in New York City. And it was something where I, I’ve spoken about before, but I have a autoimmune disease. And so I kind of knew I couldn’t do like, years and years of teaching because even the, the, the three years that I did as a teacher kind of was maxed me out in terms of my health. But it was something really, like, important for me to like, I always wanted to help people and I really enjoyed teaching. And then getting thrown into this crazy world where I legitimately was the only, you know, biology teacher in the whole school. And I had, you know, five different preps, if you’re an educator listening to this, I had literally 5 preps as a baby 22 year old in one of the most complex environments in the US to teach. And so, you know, teaching science and being able to support students, especially in their first in New York City, it’s their first time to take a standardized test in high school is Deliver environment exam in ninth grade.

Amanda Bickerstaff [00:05:29]:
So it was a really like, interesting and challenging time. But like, when I left teaching, you know, I kept, I stayed in the STEM world for a long time, often on STEM plus, like professional learning to start. So always was connected. I ran a coding for kids company for a bit. We even opened like, like a, like a tutoring session, like place. I don’t know if you’ve ever, like, you know, talk about the weird experience that I have in my career. Real estate was a new one. Like I kept going and finding like a storefront and doing interior design and it was really fun.

Amanda Bickerstaff [00:06:01]:
But so I sucked. Kept in like the STEM world, but then really moved into like leadership positions in ed tech. And so, you know, that became broader. But it was always really interesting because even when I was like building, let’s say advancement courses, which is online professional learning, we really did a lot of work around building, like really strong STEM approaches, whether that was computer science education, you know, training for educators on like, problem based learning and project based learning and you know, the. We love the 5e model, like all of those pieces. So I always kind of kept connected. But then when I went to the last job I had before I started AI for education was a CEO of an ed tech that really was about student perceptions of teaching and then became, we built a wellbeing tool. So really kind of stepped back and was really just focused on the work of teaching readiness to learn all those things.

Amanda Bickerstaff [00:06:54]:
But during that process, it was the closest I’ve ever been to actually building with technology. And so we were replatforming, we were in the weeds of like, oh man, this is why AI for education has no technology. It was very, very difficult to build technology for schools, but just to build technology in general. But it was interesting to kind of take all that support and that work around STEM and actually see it in progress and see the importance of, you know, computational thinking and all this, this work that, you know, had been tangential to my career, that led me, I think, to the work that I do now, really interestingly, because taking kind of everything that I’ve learned, whether it was being a teacher, whether it was working in STEM, whether it was running in ed tech, and when ChatGPT, the first time I ever used it, and this is a true story, like legitimately, the first time I used It I started AI for Education. And the reason being is that I think all the things that I’d ever learned in that moment crystallized. And I think this is when people talk about your light bulb moment, it always is so trite. So I apologize, everybody. It is incredibly trite.

Amanda Bickerstaff [00:07:55]:
But I think the light bulb moment for most people is really like a culmination of all your experiences, plus the right galvanizing moment that leads to this. This moment where you realize what needs to be done. And for me, what I realized it needs to be done is based on everything that I learned, is that, you know, people would need help. And when I say people, educators, leaders, teachers, students, parents would need help understanding what generative AI is, of course, but also what the impact would be. It was really interesting though, because I don’t think I would have come to this conclusion A, so quickly or B, at all if I hadn’t had this whole journey of like all these different, you know, parts of the approach, but also just having always been someone that had been thinking about, like, how things fit together as a STEM educator, because I think it’s a big part of what we do. And that led me to starting AI for Education.

Rebecca Joyner [00:08:47]:
And so what would you say is like, the primary purpose of AI for education for anyone who’s unfamiliar with it?

Amanda Bickerstaff [00:08:53]:
So our stated goal is to train 1 million educators on AI literacy, which is pretty cool. And we’re definitely in a year like, you know, it’s a two year anniversary of the company in July in terms of it actually being companies. So we’re on our path. We were kind of reaching 300,000 educated already, which is pretty cool. And so we focus on AI literacy first because we believe very strongly this is a foundational literacy moving forward. As AI becomes generative, AI specifically becomes more ubiquitous and proliferates in our private lives and schools and work. It’s very, very important to be able to understand what these tools are and aren’t, how to use them safely, effectively and ethically, which is. We call this the C framework of the AI literacy.

Amanda Bickerstaff [00:09:37]:
But really the reason why we focus so much in AI literacy is that schools need to change. Schools have needed to change for three decades at least. And we’ve been very, very good at not changing. We’re very good at, you know, kind of rigid structures in education for all kinds of reasons. And so ultimately, AI literacy for us leads to school transformation because the more that you build an informed, you know, population of teachers, leaders, students, parents, the more you realize that the traditional structures that we’ve been working within. No longer work. You know, we were assigning essays back when, you know, it was private tutors and, you know, boys learned. And, you know, this is.

Amanda Bickerstaff [00:10:18]:
We’re still doing that today, and not just in, like, schools, like, as an assessment, like a summative assessment, but also, like, in interest exams and, you know, the ways in which we are measuring if a student can graduate or their placement or ability to get into the school of their choice. And so I think that that’s our ultimate goal is, like, AI literacy as a foundation, build that informed population to then start to really look at our systems, to be able to make positive change for our young people.

Rebecca Joyner [00:10:47]:
Yeah. Cause I guess what you’re saying is the purpose of an essay is no longer. It no longer serves the purpose it used to with AI as a tool that students can access.

Amanda Bickerstaff [00:10:57]:
Yeah. See, it’s interesting because this is where it gets really, really interesting, is that, like, what are the foundational skills of, let’s say, to talking to a STEM audience. A lab report.

Rebecca Joyner [00:11:07]:
Yeah.

Amanda Bickerstaff [00:11:07]:
A lab report is as much about following a structure as it is about scientific inquiry, scientific method. Right. It’s. Can kids sit down and take their thinking and put it through that? And so that is. Is a. A skill set. Right. But it’s a skill set that is becoming less and less viable as something that students will need to do as adults.

Amanda Bickerstaff [00:11:29]:
Meaning, like, more likely a student will take the inquiry that they have done and partner with an AI system over time and career to help structure, to help them, you know, do the feedback and the connections. And so it’s not that it’s not important, but, like, what is the purpose of a lab report now? Has to be. We need to look at it, like. And we need to look at, you know, if it really is about inquiry, then where is. Where are we really thinking about the inquiry?

Rebecca Joyner [00:11:58]:
Mm.

Amanda Bickerstaff [00:11:59]:
And does it have to be in that format anymore? I. I remember, man, y’, all, like, teaching, you know, high school biology in the Bronx is just a very funny experience. When you teach science, like, you really, like, the engagement is very hard. But trying to, kids, Kids get down and be like, here’s this. This lab. You did that. Everyone does. And then now you have to take this lab and put it into this formula of, like, here’s your, you know, your hypothesis and all these different pieces was very, very difficult for kids.

Amanda Bickerstaff [00:12:30]:
And it actually, I think, ended up discouraging them and disengaging them from the inquiry part of it, because they knew they would have to, like, figure out a Way to like, make this in this format. And I think that a lot of what I love about STEM is the exploration, it’s the critical thinking, it’s the innovation. So I don’t know, it’s been really interesting and I think it will continue to be really interesting because we’ve been asking kids to do lab reports for a really, really long time. And almost exactly the same format as when you and I were in school. Right.

Rebecca Joyner [00:13:03]:
And it’s almost obsolete at this point. And I think even when you said about your student population in the Bronx, like, I even think about some of the honor students I had in the past, and they were so concerned about getting it done right. That they totally missed all the inquiry and exploration. Like you’re saying, like the actual purpose of the lab to explore and test this out because they just wanted to make sure they got the right things down. And so I, I guess, I mean, at this point, it must have been 8 years ago I stopped doing formal lab reports and people thought I was crazy. Like, but now I, I hear what you’re saying. It’s almost like it makes me think about, like our teachers when we were in math and high school being like, you’ll never have a calculator in your pocket. And now it’s like, well, we do so we have to learn math differently because of what you’re saying.

Rebecca Joyner [00:13:47]:
So I think that’s a really interesting approach thinking about, and I love that idea of AI literacy. That’s something I haven’t thought of, like putting those two terms together. So I think that’s really interesting. So kind of related to that. What do you think then are the biggest misconceptions like teachers and students have around AI?

Amanda Bickerstaff [00:14:04]:
Well, I think probably the biggest. There are a couple of really big misconceptions. Number one is that like, this is a brand, like generative AI is a brand new technology and it is significantly unreliable. It is something that I think that the majority of people are either, like, not sure about this, not really going to use it yet, but I know it’s there, I’ll come, you know, or it’s like a risk, or it’s going to take away jobs and all these things and all these concerns that are real, and then you have others that are using these tools. But I think what’s really fascinating is because these tools are designed to approximate language and, or images or coding and to essentially respond to whatever query you give it. It is really fascinating to see how easy it is to trust these tools fully, like to engage with them. In a way that you just blindly trust the tools. And realistically, these tools are probabilistic models that are making stuff up all the time.

Amanda Bickerstaff [00:15:06]:
And a lot of times they can either be egregiously wrong or like very subtly wrong. And so that is something that, like, the trust issue for those that use it can be quite large. Another misconception is that, like, it’s just really easy to use and you’re just going to save a whole bunch of time and effort. When you use ChatGPT or Gemini or Claude, you’re coding with natural language, but you’re using natural language like English to code like you would JavaScript, C, whatever, you know, Java. And so it actually requires a significant amount of knowledge about prompting technique, of actually how to do this. And this is something that I think, because it seems again so easy, the interface is like you can type anything in. You can type in purple and it will give you something for purple, the same way that, you know, you can use keywords on Google, but it’s not at all like that at Google or any other technology. And if you use it in a very basic way, you’re going to get a very basic answer.

Amanda Bickerstaff [00:16:07]:
And so there’s research that shows that if you take two groups, groups of people, one, one that are very like nascent users that don’t have very little prompting technique, and one that is a group that has prompting technique and you give them an okay chatbot and a great chatbot, that the people with very little skill level get the same outputs and an okay chatbot and a great chatbot. It’s only those that have prompting technique that see the difference in quality because they are able to actually use it in a way that’s meaningful. And so it really comes down to like our user, like the way that we use it and to how good the outputs are going to be. And I think like the last misconception is that this is something that is, you know, a hype cycle going to go away. It hasn’t had a big impact yet. It’s not going to have a big impact. And I’m just going to bring up, because we’re in a science podcast, Amara’s Law, which says that we tend to overestimate the short term impact of technology and underestimate the long term impact. We are in that overestimation period, which is where that hype cycle exists, where there, the majority of the, the economic value is at the individual level, meaning that like, even if you have a company of a thousand people, the individuals getting the most out of that is where the economic value is going to be at a personal value.

Amanda Bickerstaff [00:17:23]:
But we are moving so fast into the unknown that if we stop development today, it would take decades to actually apply the generative AI systems we have today in the way that are meaningful. These tools are doubling in capacity and capability around every seven months.

Rebecca Joyner [00:17:45]:
Wow.

Amanda Bickerstaff [00:17:46]:
And so what we’re gonna. So it’s not slowing down, but just because it’s not impacting every part of our lives right now in meaningful ways, instead is kind of an annoyance sometimes. Like I don’t want my chatbot to be chatgpt or you know, I don’t want my kiosk when I’m ordering food. Cause it makes states really, that is like, it’s a hazy. Like you’re not, we’re not quite seeing what’s really happening where the majority of generative AI is used to code. We’re not just going to see like that AI is going to continue to get better and better and better, but like all technology is improving. Haven’t caught up to it yet. So true.

Rebecca Joyner [00:18:21]:
I mean, I even think about the first time I used it in the summer of 2023. And I, because I hate writing multiple choice questions. And so I was like, I want to see. I fed it. I was like, here’s 10 good multiple choice questions I’ve written. Can you me 10 more?

Amanda Bickerstaff [00:18:34]:
Yeah.

Rebecca Joyner [00:18:35]:
So bad. And I was like, you’re a stupid robot. And then I, you know, for a year I just wrote off AI because I was like, it’s stupid. It doesn’t know how to do chemistry. And then, you know, I interviewed some teachers on my podcast who had started using it. And, and so then I was like, okay, well, I’ll start kind of messing with it again. And already just to see in that short 18 month period of time where I hadn’t touched it, how much it had changed and got improved like you said. And that’s only just going to continue.

Rebecca Joyner [00:19:01]:
It’s so true.

Amanda Bickerstaff [00:19:02]:
Absolutely. And there’s nothing saying that when you use it 18 months ago, is that. No offense to you, Rebecca, you’re lovely.

Rebecca Joyner [00:19:08]:
No, no, no.

Amanda Bickerstaff [00:19:08]:
It’s been user error or you weren’t paying for the best model. Or like, do you think that when someone uses it for one or two times and it doesn’t work, they’re like, this doesn’t work. And they don’t. Because it feels so easy that it lulls you into this false sense of like, I just know how to use this. And realistically, you don’t and you’re right. Unless you’re intentionally taught.

Rebecca Joyner [00:19:27]:
I didn’t know how to, like, I didn’t know how to provide feedback to it to make it better. I just, I gave it one chance and it did a bad job. And so then I was like, I wrote it off. And so I just think, I think that’s so true, everything you said. I don’t take it personally. Don’t worry.

Amanda Bickerstaff [00:19:41]:
Oh, no. I mean, like, you’re lovely. So it’s not you. I mean, like, I. I think personally, I’ve trained more people on ChatGPT than maybe anybody in the US and so I’ve seen a lot of this. And so we actually talk about two parts of, like, prompting AI chatbots. One is the technique, right. One is, like, things like priming it, giving it a role and format, types of things.

Amanda Bickerstaff [00:20:00]:
But the most important thing is the mindsets of prompting has to be a conversation. You need to give it feedback. It’s going to be generic unless you give specificity. Like, you need to be, you know, like, you need to have direct. Like, you don’t take the first prompt, give it direct feedback.

Rebecca Joyner [00:20:15]:
Right.

Amanda Bickerstaff [00:20:16]:
That and even curiosity and, like, leaning into your expertise. Like, those mindsets are the things that unblock the capacity of these tools. But when do we talk about that in terms of technology use? Never, right? Like, very rarely. Even if you’re teaching computer science, you’re not really. You’re maybe talking about, like, creative, you know, problem solving and computational thinking. A lot of times what you’re doing is like, here’s a syntax, right? Do it. Here’s how to debug, like. But now it’s like, people are coding by.

Amanda Bickerstaff [00:20:46]:
Coding is all about, like, creativity and, and direction and, like, trial and error and, like, it’s so different.

Rebecca Joyner [00:20:55]:
We’ve treated it so formulaically and it’s not. It’s. It’s like you said, it’s so much more creative. I think that’s such a good word for it.

Amanda Bickerstaff [00:21:02]:
Yeah. And. And honestly, like, no one knows fully how they work, right? So if you. This is going to be in September, but it. We’re recording this right after the sycophantic apocalypse, which you guys, if you’re a big nerd like me, ChatGPT released a new rollout of their model very quietly, and they made a couple of changes. And so sigafancy is a part of every genai model. Same thing as inaccuracies, hallucinations. What that means is that the models are actually designed to make you happy.

Amanda Bickerstaff [00:21:31]:
They’ll do whatever you ask it to do, within limits, usually in terms of safety. And so these small changes meant that Rebecca the ChatGPT would have been like, you are the smartest. You’re getting a Nobel Prize. You are the smartest person that ever existed. How brilliant are you? But also, if you said something like relatively like, you know, controversial or you were looking for advice, it would overly agree with you even if it was damaging and dangerous. And that in itself, Right, is where, like, you gotta know, you gotta understand, like how these tools work to even understand that if it responds that way, that that’s a bug, that’s a problem. That isn’t. That isn’t necessarily the truth.

Amanda Bickerstaff [00:22:16]:
Because there’s no truth. Right. Like this, this tool is probabilistic. It’s just making stuff up. But it feels like you just said yes. I like to. It felt so much that people genuinely felt they were like, I don’t want to use this ever again because it’s just telling me I’m right. And most people don’t want to use these tools just to confirm bias.

Amanda Bickerstaff [00:22:36]:
Yeah, they’re looking, they’re looking for like ideation and feedback and creativity.

Rebecca Joyner [00:22:42]:
That’s interesting. I also, I read something recently too, and I’d love to know. This is totally side note. I’ve heard that it’s not good to be polite to it because it like, wastes energy when you’re like, hi.

Amanda Bickerstaff [00:22:53]:
Oh my gosh. Can we just, can we just talk about the nonsense?

Rebecca Joyner [00:22:55]:
I have to know.

Amanda Bickerstaff [00:22:56]:
Okay, so OpenAI has lost. He’s going to lose $15 billion this year. Okay. It is not because you say please and thank you. Please and thank you are two tokens and a like, like right now. ChatGPT every context winner is 128,000 tokens.

Rebecca Joyner [00:23:11]:
Wow.

Amanda Bickerstaff [00:23:11]:
That is a. That is not why you should not say please and thank you. That is just this, like, way for them to like, mitigate the negative and true environmental impacts of these tools, which traditionally happen when training a model. Not in terms of use. You shouldn’t use please and thank you because it doesn’t really make a difference. In fact, it can make both a positive difference and a negative difference. So in some queries we say please and thank you. It’s better.

Amanda Bickerstaff [00:23:36]:
In some other ones it’s not. We suggest just using direct, just using direct feedback. Like, I didn’t like this. Like, the more direct you are, the better off you’re going to be. But these tools also, even with the memory that was launched, it’s not going to be Something but A is going to have an enormous environmental impact if you decide to do it. But also this kind of statement that, like, oh, but when the bots take over, they’re not going to work. No, they’re not. That’s not how these tools work.

Amanda Bickerstaff [00:24:01]:
It’s not like, you know, the memory even isn’t really memory. All it’s doing is picking up and storing these statements over time. Like it does not remember every chat you ever had. And so, like, it just is a very silly thing, but it is, like, really interesting because the people that are running these tools don’t know what they’re doing either. Right. Why would this even help? It doesn’t help. And yet they’re building these tools like they didn’t know that GPT4.0 was going to have this new release and they wrote this whole thing about it, and they just did a really crappy job of understanding how these tools work and how to them.

Rebecca Joyner [00:24:38]:
Okay, so that’s helpful. That’s a misconception cleared up too. What would you say then, with your experience with AI? What do you think are, like, some of the best uses for science teachers and then also science students?

Amanda Bickerstaff [00:24:50]:
Well, first of all, we have a whole prompt library on our site. So if you ever want to, like, get a really good bank of, like, good first prompts or techniques, like, we have, you know, inquiry lesson plans, we’ve got STEM experiments, we’ve got all kinds of things. I think for science teachers, there’s a couple of things. One of the things that’s hardest for us is explaining complex topics in meaningful ways to young people. And so the ability to go and find like, and use generative AI to help you identify a really creative way to discuss, let’s say, the life cycle in a way that kids care about more than like a butterfly. Like, where’s ways in which to like an urban population versus rural population or even a population that has a very specific animal or, you know, flora, fauna that you have available. I think that for STEM teachers, it’s a great way to like, get experiment ideas and like, inquiry ideas and, you know, icebreakers. And I mean, just sometimes we get stuck as, as STEM educators, we have like, the one thing that we know that works, but then we get kind of stuck in terms of like, how do we engage kids in this? There’s like a hundred ways to teach mitosis and meiosis, and some of them are better than others.

Amanda Bickerstaff [00:26:02]:
But, like, what happened? Why don’t we just take a really creative approach to being like, this is how I’VE always done it, but this is the issue I see. Give me some inspiration. And often I will say that like the inspiration, you’re like, I hate 90% of this, but that 10%, oh man, I have never thought about doing this. You can, you can put into a generative IO model, here’s all the materials I have. Create an experiment or here’s a goal I have. And like you always have to edit and refine and use your expertise. But I think that those are really great options. One other piece is that I think is always going to be really interesting as a STEM teacher specifically is like, we have to do a lot of assessment, a lot of formative and some of assessment.

Amanda Bickerstaff [00:26:39]:
So they talked about the multiple choice questions. Well, like, how do we take like something and not to a multiple choice question? How do we use formative like exit tickets in meaningful ways? How are we using spot checks? How are we using these kind of other opportunities to assess like generative AI is going to be a great thought partner for that. And then I think for students, students can get feedback. Giving them AI literacy training that lets them understand the break point between and AI taking over and AI being a tool, an augmentation tool or a system tool is really important. But once you’ve done that, especially kids over 13, having them get feedback on an idea, having them, you know, test out and model, you know, like, you can create games, you can model, like, okay, so this is my materials for my egg drop challenge. Will this even work? Okay? If not, why not? Let’s talk through it. What if I did this? What if I did that? And then go and do it just so that the kids that are maybe like really concerned about always being right or working, like get very discouraged by STEM as they get older, especially call out my our girls. That’s one of the research fact reasons why they drop out of STEM in high school.

Amanda Bickerstaff [00:27:45]:
Things like that can be really, really wonderful. Also just on demand support, like, we’re not always available. You know, there’s a funny thing that Most kids use ChatGPT between 11pm on 2am we’re not available and air sets in the middle of the night. Some of these kids don’t have, you know, a lot of our kids don’t have tutoring support or other, other support at home like those with the right type of AI literacy. Even today, of course, taking with a grain of salt due to like the. We talked about the fact that it makes things up all the time can be really, really helpful. And even teaching kids to critically evaluate and understand what’s right or not or what they want or don’t want can be really powerful for a STEM classroom.

Rebecca Joyner [00:28:26]:
Yeah. I love the idea too, like you said of like, when we’re having them design an inquiry investigation about a specific topic of saying, okay, like, tell them what are these things that I have in my classroom? We’re learning about feedback loops. And you need to design an investigation where we test a feedback loop, get some ideas. And like you said, like, yeah, maybe 90% are bad, but you only need one if you’re coming up with one lab.

Amanda Bickerstaff [00:28:45]:
Oh, totally. And this is the thing, like, it could give you a thousand.

Rebecca Joyner [00:28:49]:
Yeah.

Amanda Bickerstaff [00:28:50]:
If you really just are not feeling jazzed about something, then keep, keep prompting and AI sometimes never going to be bored or angry at you for asking the same question ten times. Adult will. Yeah, my would be annoyed.

Rebecca Joyner [00:29:03]:
Right. Okay. So what do you say to the teacher, though, who’s so nervous about students cheating with AI? Because I think that’s a huge fear that teachers have.

Amanda Bickerstaff [00:29:11]:
Yeah. And I think that it’s never been easier like, to like, we. We say there’s like an easy button for every traditional assessment. So it’s not that it’s not a risk. And the research shows that kids are just mostly, it’s kids cheating differently. The kids that were cheating with crib notes or taking someone else’s paper or an SAML writer are now using generative AI. But I do think though, that why. Why do kids cheat? Because ChatGPT did not start the cheating behavior.

Amanda Bickerstaff [00:29:43]:
Like, if anything, it just exposed how much cheating was happening in a meaningful way. Because kids are using these tools in silly and, you know, poor ways where it’s very easy to, in some cases, although I will tell you, kids are getting, you know, the kids that know how to use it, it’s going to be almost impossible to catch them because AI detectors are not reliable. Underline, exclamation point, whatever we need to do. But I think that, you know, we need to kind of take a look at why we’ve seen academic integrity become such an issue as in the. Over the last couple decades and start looking at like the assessments themselves. So that’s one. And so I think that has nothing. I think that.

Amanda Bickerstaff [00:30:23]:
I think ChatGPT is a forcing mechanism to looking deeply into assessment practices. I think the bigger issue that’s more broadly, I think going to have an impact versus academic integrity is cognitive offload. I think that students not getting AI literacy work and not understanding that if I give it, like, as we talked about before, if it’s replacing all my thinking, if I’m giving it all away and I never learned those foundational skills or those application or higher orders bloom skills and that could damage me from the future because I won’t even be able to use AI systems well because I won’t be able to evaluate outputs or be able to direct outputs because I don’t know enough about this. I think that that to me the academic integrity conversation is very important. But it does. It’s not. It’s not really the issue right now where like suddenly everything that was uniquely human that we assess can now be done by generative AI with one to five prompts. And so I.

Amanda Bickerstaff [00:31:20]:
That’s where. How are we going to instill to students a what the purpose of different types of learning are? Like, instead of just saying here’s the thing you have to do, like, why is going to become more and more important than ever before. So students actually are building true knowledge of why they’re doing things and are motivated to do them. I think that second, we’re going to have to reevaluate how we teach kids in general with this new normal is coming down. And third is that it has to be something that is way more focused on durable skills on like understanding how to like build capacity and understanding informational fundamental skills, but then how to apply them in digital frameworks. I think that that’s going to be very, very important. There’s a chance that like one of your students that really wants to be a bio, you know, in biomedicine and goes. Wants to be in a PhD mph or whatever and may never actually work with live cells or aligned medications.

Amanda Bickerstaff [00:32:17]:
They could be doing everything in 10 years through modeling.

Rebecca Joyner [00:32:21]:
Which is crazy.

Amanda Bickerstaff [00:32:22]:
Yeah. And like, like meaning that like no more lab mites potentially until the very end. Like which, like, you know, ethically is like a very interesting question, but like, and it may be a positive one, but like that is really where the world is going. Right. Is that the ways in which we work are going to change so dramatically that what we’re teaching kids today, like, how are we going to ensure, or at least try to ensure that the skills still matter when it matters most to them, which is going to be when they need to have a career and in life and be happy and right. All those important parts of being a teacher.

Rebecca Joyner [00:32:58]:
It’s so true. I love what you said about at the beginning of like, I’ve said this all along. Like cheaters are going to cheat. They’ll always find a way. It used to Be, you know, writing things on the inside of a water bottle label like, and then they’re drinking the water.

Amanda Bickerstaff [00:33:10]:
Does gel pen or.

Rebecca Joyner [00:33:12]:
Yeah, they’ve always gotten creative like you said. And, and at the end of the day it’s like, like you said, it’s two parts. It’s an integrity issue. It’s not the, it’s not the tools problem. It, it’s like this is a like culture problem we have to zoom out on. But then also, like you said, it is an opportunity though to assess your assessments and say, okay, if this is actually like a really. If it’s assessing the skills we want it to like for instance, in a lab, like, yeah, the AI can help with like the lab write up. But the actual, like you said, explorative process of like doing the lab and getting data the computer cannot necessarily do for them with their hands.

Rebecca Joyner [00:33:44]:
So it’s a good opportunity to kind of reevaluate and reflect here. So you already said about the prompts on your AI for Education website, but what are other ways that you all serve teachers at AI for Education?

Amanda Bickerstaff [00:33:56]:
Well, we have a free course that is designed to be two hours and it is got a certificate. And so I just. If you really are thinking about building your own AI literacy and it goes through like educational use cases, it goes through planning use cases. We’ve got a whole bunch of free webinars. So we have a webinar actually. Well, it’ll be late, but you can check it on our website because it says the future, the future us will have a webinar just this week on a new resource on math teaching. Like so actually using AI around math, which is crazy because ketchup T is really bad at math. So we’re actually going to talk about that.

Amanda Bickerstaff [00:34:30]:
But we’ve got a really cool guide of that. So lots of free webinars. We’ve got a free curriculum. If you want to do AI literacy for your young people that you work with, it’s only four lessons. So it’s really, really simple. And then we have these, we have courses as well around instruction and pedagogy and even just Gen AI literacy training. So we do, we try to do a lot and then we work directly with schools and districts around professional development and workshops. But I mean, our ultimate goal is like we want to be at all parts of this where we want to be.

Amanda Bickerstaff [00:34:59]:
Whether you’re an individual that’s building your own AI literacy wants to do that for your students. We’ve got all of those resources. If you’re Someone that’s thinking about other people. Like, you’re a department head, we’ve got resources for you. And then if you’re thinking about bringing it in longitudinally, we got resources for you. Because we do believe this has to be from every angle. Like this rival technology that’s becoming more and more ever present in our lives. So it can’t just be like one way.

Amanda Bickerstaff [00:35:24]:
Like, we have to be able to support this work from all angles.

Rebecca Joyner [00:35:27]:
I love that. That’s so exciting. And I’ll link all these different things you’ve referenced in the show notes for anyone’s listening. And before I let you go, I have to ask two things. One, best way for people to find you other than your website. Do you. Are you on Instagram?

Amanda Bickerstaff [00:35:41]:
Oh, man. I’m on LinkedIn, everybody. Oh, gosh. So I had. Can we just talk about this? I had not posted on LinkedIn for a year before I started AI for education. And then now I’m like, it’s like a second job. Like, we’re in like four to five times a week. And so we’re very on LinkedIn.

Amanda Bickerstaff [00:35:58]:
Great.

Rebecca Joyner [00:35:59]:
Everyone always has one place that they tend to have.

Amanda Bickerstaff [00:36:02]:
I’m not a social media person, so I’m, I, I can only do one because we’re like, oh, we should have Instagram and Tick Tock and like, we’re just lucky I do one.

Rebecca Joyner [00:36:09]:
Yeah, totally. That’s how I feel. Because my only place I really exist is Instagram. Like I have a Facebook page for the sake of having one, but so I get it. So you’re in LinkedIn. We’ll make sure we link your LinkedIn. And then now you have to tell us, since the Secondary Science Simplified podcast, what’s one way that you’ve simplified your life recently?

Amanda Bickerstaff [00:36:25]:
Well, I just moved. Ooh. I no longer live on a fifth floor. Walk up.

Rebecca Joyner [00:36:32]:
That’s exciting.

Amanda Bickerstaff [00:36:33]:
So everyone, a hundred stairs, almost a hundred stairs to my apartment. So there are days where I’m like, do I. You have to be really convinced to go outside or to go down the stairs because it’s, it’s really aggressive. So I think by moving to a place with a, a much nicer place with a elevator has I, like, I go outside a lot more. I already took a walk in the rain. So I think that that was, you know, that that small change is helping me a lot in my mental health and well being. That’s awesome activity. My step tracker as well.

Rebecca Joyner [00:37:05]:
Yeah, seriously, you’re not going to get as many, you know, flights of stairs on your Elbow.

Amanda Bickerstaff [00:37:11]:
I, I won’t take. I will rather do a like this morning 30 minute walk in the rain than 15 bites of stairs. This is the funny part is that like, I live in New York City, I run a company. I like, I kind of like live in this world where, you know, I don’t cook a lot. And so I wouldn’t even want to get like, even like Uber eats or, or God forbid, like something heavy because the delivery people would be mad at me. They would like literally by like the third flight, they’re like, like literally mad at me as a human on the fifth floor. And so to not annoy delivery people and make because they have a very hard job and like, it’s thankless is also a way that I feel very positive about my simplifying your life.

Rebecca Joyner [00:37:54]:
I love that. That’s so funny. It’s just funny to think about how, like you said, one change has had so many different.

Amanda Bickerstaff [00:38:01]:
Yes. And annoying less delivery people.

Rebecca Joyner [00:38:03]:
Yes. You’re just contributing to society better that way. Well, Amanda, thank you so much. This has honestly been a treat and I’m just really excited for how listeners are going to learn from you and excited that they. That you. I’m excited that AI for education exists. So thanks for being here.

Amanda Bickerstaff [00:38:18]:
Well, we love it. We love all our send people and thank you for taking the time with us today. Thanks everybody.

Rebecca Joyner [00:38:24]:
Thank you so much for listening to today’s episode and my interview with Amanda. You can find all the links mentioned in the show notes like her website, her free course, ways you can connect with her. All of that is available at it’s not rocketscienceclassroom.com/episode201201 and if you are listening to this podcast and you’ve never left a review and you’re ready to either use AI for the first time or try to increase your AI literacy for you and your students this year. Then I would love it if you would take the time to leave a review for the podcast. All right, teacher friends, that wraps up today’s episode. If you’re looking for an easy way to start simplifying your life as a secondary science teacher, head to it’s not rocketscienceclassroom.com challenge to grab your classroom reset challenge. And guess what? It’s totally free. Thanks so much for tuning in and I’ll see you here next week.

Rebecca Joyner [00:39:20]:
Until then, I’ll be rooting for you, teacher friend.