# AI, ChatGPT-3 & Tableau - here’s how you can use all three in harmony TODAY!

> This is content from just-tim, the data-and-analytics channel by Tim Ngwena (formerly 'Tableau Tim'). Tim has 12+ years of hands-on BI experience and covers Tableau most of all, plus Power BI, Looker, Hex, SQL and data modelling, the analytics industry, and the craft of doing the job — always tool-agnostic and honest about the trade-offs.

- **Author:** Tim Ngwena (just-tim, https://just-tim.com/about)
- **Published:** 2022-12-30
- **Format:** Video · 19 min watch · transcript available
- **Topics:** AI & ML, Industry trends, Data visualisation
- **Tools:** Tableau (ai, ask data, explain data)
- **Canonical:** https://just-tim.com/posts/ai-chatgpt-3-tableau-here-s-how-you-can-use-all-three-in-harmony-today
- **Watch:** https://www.youtube.com/watch?v=mhN_hFVUgsQ

I share my thoughts on how ChatGPT, GPT-3 and Tableau can work together today. I demonstrate using ChatGPT as a conversational assistant for Tableau questions, walk through where it falls short on complex tasks, review Tableau's own AI capabilities, and argue that the real opportunity is using AI to help people discover and use the tool better.

## Key takeaways

- ChatGPT is excellent as a conversational assistant for simple Tableau questions and follow-ups, but the more detail it produces on complex topics, the higher the risk of factual errors you'd need expertise to catch.
- Tableau frames its AI under three areas: augmented analytics (Ask Data, Explain Data), business science (forecasting and predictive modelling, formerly Einstein Analytics) and data science (connecting your own Python/MATLAB models via the Analytics Extensions API).
- ChatGPT works by combining GPT-3 with large-scale human feedback over a corpus scraped from the internet around 2021 — it builds a statistical model of language, not a knowledge graph.
- Training a model on a specific corpus, like Tableau's version-specific help docs and community forum answers, could surface better ways of using the product rather than just better insights.
- If you publish blogs, consider adding a copyright notice now, as AI tools are currently scraping content freely before legal frameworks catch up.

## Chapters

- 0:13 What ChatGPT is and basic Tableau questions
- 1:52 Follow-on questions and conversational depth
- 3:10 Where it falls down on complex tasks
- 4:39 AI inside Tableau today
- 8:37 How GPT-3 was trained
- 9:35 Training on Tableau help and community data
- 10:54 Helping people use the tool better
- 13:09 Copyright concerns for content creators
- 14:12 Why this is most useful for learning Tableau

Watch the full video, read the transcript and use chapter deep-links on the page: https://just-tim.com/posts/ai-chatgpt-3-tableau-here-s-how-you-can-use-all-three-in-harmony-today

---
just-tim — Data and analytics, with a point of view. · https://www.youtube.com/channel/UC7HYxRWmaNlJux-X7rNLZyw · https://twitter.com/TableauTim · https://www.linkedin.com/in/timngwena