The implementation of AI applications relies on computer deep learning techniques and training large language models. Taking ChatGPT as an examples, AI models are bringing about profound transformations in various fields, impacting how individuals learn and work. To prepare for the era of digital intelligence, we invite all XJTLUers to explore the potential of AI applications in enhancing learning and work.
What is prompts?
Prompting is used to instruct an AI to do a task. The better prompts (ask questions and inputs) you use with generative AI, like ChatGPT, XIPUAI, the more relevant and meaningful responses (outputs) you'll receive! But it’s very important to keep in mind that AI-generated content can be inaccurate, misleading, or entirely fabricated, so be sure to carefully review any work containing AI content before you use.
Tips
01 Providing context and be specific
AI is a machine. It is not capable of reading our minds. Give it as much context as possible so that it understands the conditions under which we want to obtain our answers. Be specific about the instruction and task you want the AI to perform. The more descriptive and detailed the prompt is, the better the results.
02 Role playing
Asking the AI to behave as if it is a type of person, process, or objectcan be an easy way to start generating better prompts. The AI will attempt toemulate that role and answer accordingly. Example:Act as an expert AcademicLibrarian…Act as a high school Math teacher…
03 Consider audience and style
Give the AI specifics about who your audience is and what sort of tone you’d like to set, it will be appropriate for the reader’s knowledge level. Indicate the preferred style for the content, e.g., formal, casual, educational, etc.
04 Format the result
Generative AI can produce many different types of outputs, including code, reports, summaries, a numbered list, a table with content, markdown format etc. Being specific about the format of output you want will produce better results. After describing what you want, try adding example: Present this in the form of Markdown style; Create a table that contains the above topic.
05 Limit the length
Limit the length of your output results, indicate AI how many results you want to have.
Example, Please provide me a list of 10 topics; Please write within 300 words.
06 Build on previous prompts
It will remember what you've said in the conversation, so you don't have to repeat everything again. You can try starting with a basic question and adding to it over time. Add more context and specificity to guide the AI toward the output you’re looking for.
07 Use examples
Using an example of a sentence or paragraph that you’d like your output to resemble can give the AI a reference point to help improve results. This can help you get a better sense of what works and what doesn't, and can provide inspiration for your own prompts.
08 Include the keywords
To help the AI generate more relevant ideas, it can be helpful to include specific keywords in your prompts. For example, if you are writing about computer, you might include keywords such as "CPU," "Memory," or " Graphics card."
CONNECTED PAPERS
CONSENSUS
ELICIT
TOOLS FOR ADVANCED LEARNERS
DISCUSSING AI
Generative AI :Potential Benefits
Limitations & Ethics of Using Generative AI
Bias and Fairness
AI can be subject to bias, depending on how it's trained and what data it's trained on. For instance, an AI tutoring system could potentially favor certain types of students over others, based on the data it was trained with. This could perpetuate existing biases and inequalities.
Accessibility and Equity
The requirement for payment to access quality AI tools creates a divide based on affordability, and not all students or schools have equal access to the technology and infrastructure needed for AI-based education. This digital divide can exacerbate educational inequalities.
Privacy and Data Security
AI systems in education often rely on collecting and analyzing large amounts of data about students. This raises questions about how this data is stored, who has access to it, and how it is used. There are risks of breaches of privacy and potential misuse of data.
Reliability and Transparency
AI tools may not generate truthful or reliable answers and often lack the ability to credit the sources of their information, leading to challenges in verifying the accuracy and origin of their responses. To uphold transparency, it's important for students, parents, and educators to understand how AI tools make decisions.
Regulation and Policy
The rapid advancement of AI presents challenges in developing adequate legal and regulatory frameworks to ensure accountability, consumer protection, and fair competition. AI Policy-making in Education is complex and needs to strike a balance between protecting students and enabling innovation.
Environmental Impact and Over-reliance
Training and running large-scale generative AI models can require significant computational resources, leading to high energy consumption and environmental impact. Overreliance on AI-generated content might lead to the degradation of human skills.
Intellectual Property and Ethical Considerations
The use of existing content by AI raises intellectual property concerns, and there's a risk of AI being used for unethical purposes, such as creating deepfakes or spreading misinformation.