What is Generative AI

• Job loss, economic shifts, income inequality
• Over-reliance on AI, diminishing human capabilities
• Digital divides between those who have AI knowledge and access and those who do not
• New risks for civil and political rights as AI facilitates data collection and mass surveillance
• Global AI power imbalances as nations with advanced AI exert political, economic and military dominance over less developed countries.
• Loss of human control over AI systems

The development and use of AI systems is driving up energy demands, carbon emissions and water usage, particularly for cooling data centers. While researchers and companies are exploring ways to make generative AI more sustainable, it is still important to consider whether your use of AI is worth the environmental impact and to use generative AI tools as efficiently as you can. More on the environmental impact of AI Explained: Generative AI's environmental impact

AI models will have data that is biased, reflecting stereotypes, prejudices, and discriminatory views. Generating this type of data causes harm and perpetuates inequality. AI output may also reflect the information shaped by various political, cultural, and corporate perspectives. It is up to AI users to recognize biases and not perpetuate false or misleading narratives and stereotypes.

AI models are frequently trained on copyrighted works without the creators' permission. This raises legal and ethical questions about ownership. There is a risk in generating content that is very similar to copyrighted works without proper authorization. Legal challenges are taking place and will certainly influence copyright laws and institutional policies. We should not assume it's free to use without checking the rights on the original source.

Sharing personal information with AI systems can pose risks to both you and others. Such data may be accessed or repurposed by external entities, including government agencies or private corporations, often beyond your control. These concerns are particularly acute in authoritarian contexts, where surveillance and political oversight are prevalent. To mitigate these risks, review and adjust AI system settings to prevent your data from being used to train future models. Additionally, avoid disclosing information about colleagues, students, or others without their explicit consent.

While many students are initially interested in using chatbots, the novelty can wear out pretty quickly, especially when they're surface-level interactions (Deng & Yu, 2023; Wu & Yu, 2024). These surface-level interactions where the responses seem vague or repetitive frustrate students and can cause them to disengage.

At this point, students may not use chatbots to learn more, but they could still rely on them as a shortcut to do their work for them rather than working through problems or thinking critically about the material (Aad & Hardey, 2025; Abbas et al., 2024).

Chatbots can also take the place of meaningful conversations with instructors and classmates, which are often what keep students interested and involved in their learning over time (Aad & Hardey, 2025). Without that human connection, students may begin to disengage and, again, only rely on the chatbot for answers (Abbas et al., 2024).

• 2025 Guide to Preventing Chatbot Cheating | Higher Education
• https://hai.stanford.edu/news/ai-overreliance-problem-are-explanations-solution
• https://www.sciencedirect.com/science/article/pii/S0747563224002206
• https://slejournal.springeropen.com/articles/10.1186/s40561-024-00316-7

Other Risks and Limitations

• Hallucinations: Generative AI systems can sometimes produce false content that looks credible. It makes sense in the pattern of data but not in reality.
• Inaccurate sources: If the training data contains inaccuracies, the bot may replicate them. This can lead to more misinformation.
• Outdated information: Many Generative AI models are not trained on data of current events. It might generate information that is obsolete, leading to misinformation.
• Adversarial Use: Malicious actors can exploit generative AI for misinformation, impersonation, scams, or cyberattacks.
• Erosion of Trust: The proliferation of synthetic media (e.g., deepfakes) can undermine public trust in institutions, journalism, and democratic processes.
• Lack of Explainability: AI systems often operate as "black boxes," making it difficult to understand or justify how a particular output was generated.

Read more about 6 Risks of Generative AI & How to Mitigate Them in 2025

GAI can help you automate mundane tasks and free you engage with different ideas. It can be used a collaborator stimulating creativity in your process.

By generating prompts to learn skills that are specific to your interest and level of understanding can enhance your learning for courses and the workplace.

specific generative AI tools can help reduce the time of the initial search curating articles and data (Elicit, Consensus and Inciteful) and reviewing the content more quickly (TLDRthis and AskYourPDF)

Generative AI tools can provide personalized and adaptive learning experiences. Many educational technology software and apps already include Generative AI and are widely used in k-12 schools. It may analyze students' strengths and weaknesses, tailor educational content accordingly, and offer real-time feedback. These systems can also generate interactive exercises, quizzes, and simulations to reinforce learning. University students can benefit from this personalized approach, receiving targeted support that caters to their individual needs and learning styles.

Generative AI helps with language learning and translation. It uses advanced models to create natural-sounding text. Students can use AI-powered platforms to practice speaking, writing, and understanding different languages. Also, AI translation tools offer accurate and quick translations, making it easier to access educational materials in different languages. These tools not only improve language skills but also encourage cultural exchange among students from different backgrounds.

Generative AI can offer a variety of support materials and interaction methods tailored for students with disabilities, neurodiversity, multilingual backgrounds, and other challenges they may encounter. Tools with natural language processing and speech recognition can help support students with visual and auditory needs. Involving individuals with disabilities must be at the forefront of these tools' development to ensure accessibility and inclusivity

• Elicit and TLDRthis