Accountability for students - students in my coding classes can use AI to complete their problems for them. I minimize this by having them write their code in codeHS where I can see their version histories, this can be used to show pasted code as well as whether they spent time thinking about the solution and what they were writing or simply copying from somewhere. I also have them explain code, change it, and re-explain the new scenario.
Plagiarism and academic dishonesty, Accountability for students, and Overreliance and loss of critical thinking are my top 3 issues with AI. I have been able to try to overcome these by using AI (MagicSchool) to create AI-proof assignments/projects. While theyâre not 100%, its better than a prompt/essay response. As Iâve stated before, most kids will work harder trying to cheat than to just complete the assignment and use critical thinking.
My biggest concern is that students will not learn necessary skills if AI is introduced to students too early. In the past with computer related advances, students are the first to understand and utilize the potential and teachers are the ones playing catch-up. At this stage we maybe have a very slight advantage in that our generation of teachers are already very computer literate. We also have classes that help us to quickly get on the upside of the learning curve on this. But as a student, one of my very first experiments when I discovered what it could do, was to see how well it could do it. I went into Khan academies Khanmingo and I had it generate a research paper for me as an experiment. I was really alarmed with what I saw. Not only did it generate an extremely well written college level paper, it also cited its sources and it used a lot of sources, many of which I didnât have access to. The voice of the paper was taken right from the style of the books where it got itâs information, so it sounded very beautiful and almost romantically lyrical. You couldnât really tell that it was AI from the language it used. The real give-away that could tell, was the sheer volume of resources, there was no way I had access to so many resources and certainly no way I could have read all those books in one semester and extracted just exactly all the right information. When you think about it, the time saving benefit that I would approve of, would be to provide students with that research more efficiently. I donât see much value in students sifting through mountains of research that doesnât apply to their topic. So perhaps if we could program AI to provide students with more efficient research, then students would be able to leverage AI to learn exactly what theyâre interested in and the time they use would be in writing their paper and distilling down that relevant information. In this way, learning to formulate questions would be a form of critical thinking, that would help them find what they need for their project. They would still need critical thinking skills to do that task and to write their papers, but their time wouldnât be wasted at the library and reading every book they can find on the subject. For more obscure topics, libraries donât always have the books you need, and purchasing them yourself becomes and equity issue. In many ways, a more advanced and refined search engine is what students need to be successful.
I enjoyed reading about your experiment. That is very scary that AI generated a very well written paper. I like your idea about programming AI to provide students a more efficient research.
AI is a tremendous aide in formulating difficult-to-write emails for parents when you are addressing challenges with students in the classroom. Without the emotional attachment of potentially inflammatory situations, it can provide diplomatic language and a structure that might not have occurred to me. I have used it to let parents know about low grades and truly negative behavior choices. While I have edited the suggestions from AI, the email was much easier to write in a timely manner without parents feeling attacked or their student being disparaged.
Writing assessments for enrichment classes is still somewhat of a challenge because I am the only person who teaches my subject areas in our school (theatre arts AND computer technology). AI has been extremely helpful in some ways, but I am still struggling with narrowing down the details to match standards for younger grade levels.
Deciding on what to use the LLM for to support my students in an academic task related to STEM was a challenge, but it was successful⊠that chatgpt was able to offer a task that I really probably found useful, and felt the reflection prompt which I liked most could be adapted
I chose accountability. As a Montessori teacher in an Innovation Lab, I want students to think for themselves and take responsibility for their work. But I see that using AI can sometimes make it too easy for them to let the tool do the thinking. This can take away from real learning. To help with this, I will teach students to use AI as a helper, not as a way to skip hard work. Iâll ask them to explain their choices, talk about how they used AI, and show what they did on their own. This turns a problem into a chance to build thinking and honesty.
Plagio y deshonestidad acadĂ©mica, considero que uno de los mayores retos es alfabetizar a los estudiantes frente a los dilemas Ă©ticos que implica el uso responsable y consiente de la IA como una herramienta que facilita los procesos formativos y no como una herramienta que lleva a la mediocridad y el facilismo. Es necesario acompañar el proceso y alentar a los estudiantes con propuestas que favorezcan su autonomĂa y sobre todo su honestidad.
One concern I have is student accountability. When students do not take responsibility for their writing, there is no learning. To combat this lack of accountability, I can set clear guidelines, establish explicit rules on acceptable AI use, and require citations for AI-generated content. Another idea is to require students to demonstrate core skills without using AI before integrating it into assignments. AI could be used as a learning partner for tasks like brainstorming, drafting, or feedback.