Generative AI & LLMs
Foundation model
A broadly trained model that can be adapted to many downstream tasks.
Example
One broadly trained model is adapted for summarization, search, and customer support.
Why it matters
It lets teams build specialized tools without starting all learning from scratch.
You'll hear it
“We're adapting the same foundation model for a different job.”
Can you control it?
Set by the team
The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.
Common questions
- Does broad training make every specialist answer reliable?
- No. A particular profession or task can require knowledge and checks the starting system lacks.
- Can two products using the same foundation behave differently?
- Yes. Later training, instructions and added tools can change their behavior.
- Is a foundation model always a language model?
- No. Broadly trained systems can work with pictures, sound or other kinds of information.