AI and Universities

As artificial intelligence is reshaping industries, how will universities prepare students to be needed in the job market? As I think about where I want to study and what I want to study, here are some of the ideas I am reading and thinking about:

1. There are limitations to what AI can do, and the first step will be for students to understand how AI works and what it is good at and not good at. Students need to understand how AI is getting to its outcomes and be able to evaluate the accuracy of those outcomes. In my high school, there is very little AI in the curriculum. Teachers are trying keep students from cheating with AI. However, the reality is industries are using AI, and students must learn how to use AI responsibly.

   
   

2. How can universities help students to contribute when AI is able to do many tasks that highly-trained professionals used to do? Universities will need to help students differentiate themselves. Human judgment and connection will be critical skills in the workforce. Students need to be able to offer new, creative thinking and contribute human elements like ethics and empathy. 

   
   

3. Finally, universities and students will need to be adaptable as the job landscape keeps changing. Those universities that make changes first will have a huge advantage in attracting students to enroll and employers to hire their graduates. 

   
   

   

   Source: https://www.vecteezy.com/photo/23030611-artificial-intelligence-ai-business-people-showing-future-technology

Sources: https://www.pbs.org/newshour/show/how-the-next-wave-of-workers-will-adapt-as-artificial-intelligence-reshapes-jobs

https://www.forbes.com/sites/williamarruda/2025/06/08/how-to-stay-ahead-of-ai--the-human-skills-universities-must-teach/

President Trump’s Executive Order on AI and K-12 Education

On April 23rd, President Trump released an executive order for “Advancing Artificial Intelligence Education for American Youth”. The President hopes to better prepare American students to compete with students from other countries as the impact of AI grows. However, the President is simultaneously cutting funding for education. How will teachers become knowledgeable about AI without substantial investment made for their training? Schools need help from outside partners to understand the technology, to decide how to best integrate AI education into existing curriculum, and to evaluate how to handle risks they may face. 

Here are some questions I think need to be considered:

Will there be a decline in critical thinking if students rely on AI? What will be lost in the learning process if students rely too heavily on AI?

Will there be data security issues with more wide spread AI use? How will data sources be protected and validated? Are there cybersecurity threats?

Will we lose critical human interaction? How will that affect mental health?

Could AI models be trained on biased data?

My high school has a task force with administrators, teachers, and students discussing these types of concerns. However, the technological innovations are outpacing the discussions, and teachers and students are left without clear guidelines on how to best teach/learn in a world with powerful AI tools at everyone’s disposal.

Source: https://www.whitehouse.gov/presidential-actions/2025/04/advancing-artificial-intelligence-education-for-american-youth/

Sources:

https://www.edweek.org/technology/what-trumps-draft-executive-order-on-ai-could-mean-for-schools/2025/04

https://www.whitehouse.gov/presidential-actions/2025/04/advancing-artificial-intelligence-education-for-american-youth/

AI - Model Collapse

Source: https://www.forbes.com/sites/bernardmarr/2024/08/19/why-ai-models-are-collapsing-and-what-it-means-for-the-future-of-technology/

Despite how advanced AI has become in recent times, it is still reliant on humans. Researchers published a paper in Nature showing that the more AI-generated data models are trained on, the worse those models perform. When AI models are taught by other AI, they lose some key information that would be learned if humans trained them. When this happens the experience a problem called "model collapse." Model collapse happens either in an early stage where the AI forgets about rare or uncommon things in the data, or in the late stage where the AI's outputs become more repetitive. This AI failure is important because it shows that AI is not becoming too advanced yet and that despite everything it is capable of it still needs human aid.

To avoid model collapse AI will need to be trained by quality, human data instead of being trained by AI generated content. This is hard to do because it is becoming increasingly difficult to distinguish between AI content and human content. To make sure that AI models are accurate models need to find to find diverse and authentic human data. This will require companies to coordinate efforts and resolve ethical issues around information ownership. AI is a powerful, transformative technology, but it is only as strong as the databases behind it.

Sources:

https://www.nature.com/articles/d41586-024-02420-7#:~:text=Collapse%20happens%20because%20each%20model,nothing%20else%2C%E2%80%9D%20says%20Shumailov. (July 24, 2024)

https://www.forbes.com/sites/bernardmarr/2024/08/19/why-ai-models-are-collapsing-and-what-it-means-for-the-future-of-technology/ (August 19, 2024)