Generative Artificial Intelligence

Generative AI is a type of artificial intelligence (AI) that utilizes machine learning algorithms and data sets to produce or remix content based on prompts inputted by the user. It is a rapidly evolving technology that impacts the way we create content and solve problems. Some well-known tools include the text generators such as OpenAI'sChatGPT, Microsoft's Copilot and Google's Gemini, alongside image generators such as DALL-E and other generative AI tools.

When you are using these tools, it is important to think critically and evaluate the reliability, accuracy and context of the content generated.

Generative artificial intelligence is a complicated set of technologies. While it's not necessary to have a deep knowledge of how these technologies work to use AI tools, it's helpful to have a basic understanding of some fundamental concepts.

What's Generative AI?

• There are numerous types of generative AI that can synthesize music, art, video, text, code, equations or even a blend of these.
• Put very simply, AI text generators work by predicting the next word in a sequence. They do not "understand" your questions in the way that a person would.
• AI text generators, such as ChatGPT, Gemini, and DeepSeek, are trained on a large amount of text harvested from the Internet. This training is supplemented by annotation and feedback testing done by humans (largely in underdeveloped regions or the Global South). AI image generators, such as Midjourney and DALL-E are trained similarly.
• With some exceptions, we don't have a lot of information about how the models that drive these tools were trained. That means that, in most cases, users have no way of knowing what information an AI tool has had access to. Some tools, such as Meta's LLAMA or Happy Flyer's DeepSeek have a permutation of "open box code", wherein you can see some, but not all, of the code.

Some limitations and causes for concern:

• Most AI tools capture and reuse a significant amount of user data, mostly for training models. You should not share private information with ChatGPT or similar tools. Some services allow users to exclude their data from training models, but this may not always be the case and is not on by default for many free or "low-cost" models.
• All AI text generators produce plausible sounding but false information (known as "hallucinations"), and by their nature will produce output that is culturally and politically biased. Please look at "Ethics & Copyright" to learn more.
• There are a number of ethical and societal concerns about generative AI, from the way these tools were created, to the way they are being applied now and in the future. Again, please look at "Ethics & Copyright" to learn more.
• Responsible and ethical use of genAI is dependent on your evaluation skills. Generative AI uses complex algorithms to produce highly realistic content & can be difficult to distinguish from content created by a human. This means critical thinking skills are essential to evaluate the authenticity and accuracy of what's been generated. 

"GenAI Agent" by Marxav on Wikimedia Commons, licensed under CC-BY-SA 4.0.