Generative AI Basics
Generative Artificial Intelligence is a subcategory of artificial intelligence, a field that also includes things like image recognition, recommendation systems, search rankings, and autonomous vehicles. GenAI refers specifically to tools that create new content, such as text, images, audio, video, or code, based on patterns in the data on which they were trained.
It’s more than just asking a chatbot to polish an email. It’s a powerful and evolving technology that also raises questions around originality, bias, privacy, creativity, and how we learn and work together.
| What GenAI can do | What GenAI cannot do |
|
|
How Generative AI Works
GenAI tools are trained on vast collections of text, images, and data, and use machine learning models to detect patterns within that information. When they respond to a prompt, they generate the most likely next word, pixel, or pattern. They do not understand content the way humans do; they rely on statistical relationships rather than meaning.
Impacts of Generative AI
As we explore GenAI, it’s important to recognize that these tools offer many benefits, but they can also create harm. When we use GenAI as a thought partner or content creator, we still hold ownership of the results and responsibility for their impact.
- Bias: Because GenAI is trained on vast amounts of online data, it also reflects the same racial, gendered, and cultural biases found online. In a learning environment committed to antiracism, we must pay close attention to what GenAI includes and what it leaves out; whose perspectives it centers or defaults to; and the risk of amplifying or perpetuating harmful stereotypes.
- Human Impact: GenAI depends on invisible human labor, often undertaken by underpaid workers in the Global South. It can also replace work that humans would otherwise be paid to do. We should keep both impacts in mind as we decide how to use these tools.
- Consent: GenAI draws data from many sources, often without the creators’ knowledge or permission. At CC, we are committed to understanding knowledge creation, including where ideas come from and how we carry them forward.
- Carbon Cost: Running and training GenAI uses large amounts of energy and water to power and cool data centers. Even though these impacts happen off campus, the environmental impact is real.