What is Generative AI?
Generative AI is artificial intelligence that creates new content, such as text, images, audio, video or code, in response to a prompt. It learns patterns from large amounts of existing material and uses them to produce output that did not exist before.
AI vs Generative AI
Most AI before generative AI classified or predicted: is this email spam, what will sales be next month. Generative AI produces something new: a paragraph, a picture, a summary. It is a subset of AI, built on deep learning.
How Generative AI Works
- Text: large language models predict the next word, one word at a time (see text generation).
- Images: diffusion models start from random noise and refine it step by step into an image that matches the prompt.
- Both are trained on huge datasets, and the quality of that training data shapes what they produce.
Examples of Generative AI
- Chat assistants: ChatGPT (OpenAI), Claude (Anthropic) and Gemini (Google). ChatGPT is generative AI.
- Image generators: DALL·E, Midjourney and Stable Diffusion.
- Code assistants: GitHub Copilot.
Generative AI in Academic Writing
- Brainstorming and outlining.
- Suggesting wording, for example Papero's autocomplete, which proposes the next sentence from your own papers, with the citation already attached.
- Summarising and paraphrasing sources you have read.
- Feedback on the clarity of a draft.
Risks and Ethical Questions
- False content: it can state errors and invent references (see AI hallucination).
- Bias: it repeats biases found in its training data.
- Authorship and integrity: submitting generated text as your own can breach academic rules.
- Privacy: text you paste in may be stored by the provider.
Best Practices
- Use it to support your thinking, not to replace it.
- Check every fact and source.
- Declare its use where your institution or journal asks.
Full guide: How to Cite ChatGPT and Other AI Tools: APA, MLA, Chicago