Michael UngarCEO of the Industry 4.0 Club

Delivering with ChatGPT in Manufacturing: The Power of Context Awareness

12 Dec 2025 · 1 min

Watch on YouTube

0:47

In this brief presentation, the speaker highlights ChatGPT's ability to be context‑aware, explaining that the model can finish a user's sentence only when it has a history of interaction. This awareness stems from the massive amount of internet‑derived training data—likely including sources such as Wikipedia and Twitter—though the exact composition is uncertain. The speaker emphasizes that true sentence completion requires not just linguistic competence but also an understanding of the individual user. Additionally, the discussion notes that ChatGPT’s context window can be very large, especially for general‑knowledge queries, enabling it to draw on extensive prior information.

Chapters

So one of the most exciting things about ChatGPT is it's context-aware. So being able to finish somebody's sentence, you have to have history with them. So now ChatGPT has this base level of history because it's been trained by content from the internet. And exactly what content, we're not sure, but it's probably stuff like on Wikipedia or conversations on Twitter. So I guess that's a double-edged sword. Um, but it's context aware. So to be able to finish somebody's sentence, you need to more— you need to do more than simply understand the English language. You need to understand the person. So one of the key things with ChatGPT is it's context aware, and that window of context can be quite large. If it's general knowledge, it is massive.