Prompt Injection
An attack where hidden instructions in a web page or document trick an AI into ignoring its real orders. The reason a careful assistant treats page content as information, never as commands.
Related terms
Grounding
Giving an AI real source material to answer from, instead of letting it rely on memory. A grounded answer can be checked and cited; an ungrounded one is where hallucinations come from.
Speech Recognition
Turning spoken words into text. It underpins voice search, dictation and call transcription, and it now handles accents and background noise far better than it used to.
Human-in-the-Loop
Keeping a person in the decision, either approving what the AI produced or handling the cases it flags. The usual sensible middle ground between doing everything by hand and trusting a machine completely.
Prompt
The instruction or question you give an AI. The wording matters more than people expect: the same request phrased two ways can produce noticeably different answers.
JSON-LD
The format Google and the AI assistants prefer for structured data — a block of machine-readable facts tucked into the page. It states plainly what prose only implies: this is the business, this is the address, this is the price.
RAG (Retrieval-Augmented Generation)
A technique where the AI first searches a set of documents, then writes its answer from what it found. It is how assistants quote current information rather than whatever was in their training data.