On: The new skill in AI is not prompting, it's context engineering

[__loam]

It's all the same blob of text in the api call

[chestervonwinch → __loam]

There is a conceptual difference between a blob of text drafted by a person and a dynamically generated blob of text initiated by a human, generated through multiple LLM calls that pull in information from targeted resources. Perhaps "dynamically generated prompts" is more fitting than "context", but nevertheless, there is a difference to be teased out, whatever the jargon we decide to use.

[FeepingCreature → __loam]

There's always been a distinction between prompt and data.

[simonw → FeepingCreature]

LLM's can't distinguish between instruction prompts and data prompts - that's why prompt injection attacks exist.

[FeepingCreature → simonw]

I agree, and that's a problem. It doesn't mean the distinction doesn't exist, in fact it shows the opposite.

[oblio → FeepingCreature]

Spoken like a non Lisp programmer.