Explainers

How it works, explained. (52)

What a topic is and how it works, with examples. Explainers don’t date: the news covers what is happening, this is where it gets explained.

What is function calling in AI?

Function calling lets a language model produce a structured request to run a software function, so it can use calendars, databases and APIs.

Guardrails are the rules and filters that check what goes into and comes out of an AI system, keeping it on topic and preventing harm.

AI hallucination is when a model states something false with full confidence. Invented sources and wrong dates are typical examples.

Inference is when a trained AI model produces an output for new input. Every message you send to a chat assistant triggers inference.

Influencer marketing means brands partnering with trusted creators on social media. Ads must be labeled, and results must be measured.

A LAM is an AI model built to go beyond conversation and take real actions in software and app interfaces.

Latency is the time between sending a request and getting the answer. In AI products it shapes the user experience directly.

An LCM is an experimental AI model that works with whole sentences, or ideas, instead of individual words. Meta introduced the first one.

Marketing is the set of activities that gets the right product to the right people, at the right price, through the right channel.

An MLM is a language model trained by guessing hidden words. It understands text better than it writes it; BERT is the best known one.

Google now indexes sites based on their mobile version. Here is what a mobile friendly site needs and how to test yours, step by step.

MoE splits a model into many expert sub-networks and runs only the relevant ones at each step, making large models cheaper to run.