Content:
Professional training with a guarantee of employment: "Methodologist from entry-level to professional"
Find out more
Head of the educational institution of Tyumen State University
Venue: conference "Modern {DIGITAL} didactics" at Moscow State Pedagogical University.
If artificial intelligence is introduced into the educational sphere without changing the very structure of the educational model and process, this can lead to inefficiency. In this context, AI can be compared to medicines. If you offer a person a medicine and tell them, "This is great, take it," and they decide for themselves how to use it, it could lead to a variety of errors. For example, they might take it in excessive doses, too often, at the wrong time, or in the wrong combinations, which could be dangerous, sometimes even literally. The situation is similar with artificial intelligence: without proper controls and guidelines for its use, we risk making things worse—or at least creating significant dangers.
Head of the Educational Institution, Tyumen State University
Venue: Conference entitled "Modern Digital Didactics" at the Moscow State University of Psychology and Education.
If artificial intelligence is introduced into the educational process without changing the educational model itself, it can lead to ineffective results. In this regard, AI can be compared to a medication. If a person is simply offered a new medication with the recommendation, "This is an excellent remedy, take it," but they take it at their own risk, serious errors can occur. For example, they may use it too often, at the wrong time, or in inappropriate combinations, which can ultimately threaten their health, sometimes literally. The situation with artificial intelligence is similar: a lack of control over its use can exacerbate the situation, creating certain risks.
Context
On October 16, the 7th International Scientific Conference "Modern Digital Didactics" was held at Moscow City Pedagogical University. The event focused on the application of artificial intelligence in the educational process. In a session entitled "AI Practice in Universities: What's Time to Agree," Ulyana Ravedovskaya, Director of the School of Education at Tyumen State University, presented a report in which she discussed the need to change educational models in higher education institutions in light of the development of generative AI and proposed options for these new models.
There's a clear trend in how students interact with AI-based tools, especially generative neural networks. When faced with a new task, they tend to immediately turn to AI for help without attempting to understand the issue themselves. This results in them failing to develop new knowledge and skills. Moreover, students often delegate not only routine tasks to neural networks, but also the most complex ones, which they are already familiar with. Thus, the optimistic assumption that people will only delegate routine work to AI and tackle more complex creative tasks themselves is not supported by reality—in reality, they behave quite differently.
As Ulyana Ravedovskaya emphasized, these are not just individual observations by teachers: a number of studies have already accumulated confirming these findings. In particular, it's worth noting the high-profile study conducted by scientists from MIT. The experiment revealed that participants who repeatedly used the chatbot to compose essays subsequently struggled to write their own texts—their brains seemed to have become lazy.
In the key part of her presentation, Ulyana presented an example of how the educational process at universities is being transformed. She shared information about the course "Biology of Human Behavior," which is being offered as part of a joint undergraduate program between the Skolkovo Moscow School of Management and the Moscow Institute of Physics and Technology. Experts from Tyumen State University also participated in the development of this course.
A new intensive course was recently held, using AI personalities instead of traditional teachers. Skillbox Media previously shared the findings from the previous launch, which can be found here and here. Ulyana Ravedovskaya discussed how this entire initiative was organized in more detail in her interview.
The idea is that students use an AI avatar, a virtual teacher trained on a comprehensive dataset related to a specific topic. However, students rarely experience traditional lectures; they are constantly learning a new area of knowledge, working in teams with the help of AI. They prepare reports and present them to experts, and so on.
Of course, this method has its challenges. One of them, as the speaker noted, is the significant resource investment. First, it requires extensive development time, which includes rethinking course objectives. Second, it requires active student support throughout the entire process. This increases the workload for both instructors and students. However, this approach encourages students to learn new things on their own, rather than relying mindlessly on neural networks to solve all their problems.
Also read:
- Apparently, the use of neural networks is widening the existing gap: successful students continue to progress, while those who are doing poorly begin to lose motivation.
- Danger or Opportunity? What concerns does progress in artificial intelligence raise among educators around the world?
- In Russia, about 25% of student assignments contain elements that indicate they were created using artificial intelligence.
- Experts have analyzed which types of assignments will be most effective for students in the context of the modern influence of artificial intelligence.