Note:
This feature uses an integration with OpenAI. Read below for more information on what data is shared with OpenAI and instructions on how to opt-out of AI integrations if your company policy requires this.
In this article:
About Iris conversation insights (powered by OpenAI GPT)
Iris is Brandwatch’s artificial intelligence (AI). It uses cutting-edge proprietary and integrated AI models to help you work faster and smarter. Iris conversation insights leverages OpenAI GPT (Generative Pre-trained Transformer) technology to summarize the main themes of the mentions in your Listen Saved Search dashboard charts and data points.
Generative Pre-trained Transformer technology is an advanced language model developed by OpenAI. It utilizes deep learning techniques and a vast amount of training data to understand content and generate insights. Visit Open AI’s website to learn more about OpenAI and ChatGPT.
By using Iris conversation insights, you can cut down the busywork of analyzing the themes in your mentions manually so you can focus your attention on validating, prioritizing, and further exploring those themes. We recommend using Iris conversation insights in conjunction with manual review of mention snippets, use of the topic cloud, filtering, and segmentation of your data.
Using conversation insights in Listen
When you view a Saved Search dashboard in Listen, Iris conversation insights are accessible via the right-hand Mentions sidebar. The Mentions sidebar will list mentions that open once you click through a datapoint in a component — for example, a peak in a line chart or topic in a topics cloud.
Whenever you click on a chart element or datapoint in your dashboard and view the underlying mentions, you can access conversation insights by clicking Summarize Mentions next to Iris Conversation Insights AI.
Once you open conversation insights, Iris will automatically begin to analyze your mentions.
Note:
Loading times for conversation insights may vary.
Once Iris has analyzed your mentions, it will return a summary of themes and topics in the Mentions panel.
At the bottom of the Mentions panel, click Copy to Clipboard if you would like to copy your mentions summary.
Note:
If you close the Mentions pane and reopen it, the AI may generate a new, different summary for the same datapoint. At this time due to API limitations, OpenAI may also return more text than is allowed by the API limit, resulting in a potentially cut-off response.
Opting out of AI features
- In the left-hand navigation sidebar of Social Media Management, click on Manage Settings (gear icon), then click Organization admin.
- Under OpenAI features, switch the Enabled toggle on the right side of the screen to Disabled.
Note:
This setting will disable all AI features in Social Media Management. You can return to this menu to re-enable your AI features at any time. If you have access to both Social Media Management and Consumer Research and would like to opt out of Consumer Research’s Iris Conversation Insights, you will need to opt out separately via your Consumer Research settings.
FAQs
Which model does Conversation Insights use?
Iris Conversation Insights makes direct use of OpenAI’s large language model (LLM) ChatGPT, specifically GPT 4-o-mini.
Does Iris Conversation Insights work for different languages?
Iris will summarize the content in any language, while the summary output for now will only be in English.
Does Iris Conversation Insights return the same summary for the same data point each time?
We cache the 5 summaries a user gets back from OpenAI when they click to summarize mentions from a specific dataset. If a user returns to the exact same dataset and clicks to summarize the same results again, they will receive the same summaries. By preventing the same data sample being sent over to OpenAI to be analyzed multiple times, it removes the possibility of discrepancies between each analysis. The results stay cached even if a user navigates to another part of the platform, and would only be removed if a user clears their cache or refreshes their page.
Which samples does Iris Conversation Insights send?
We are sending the 240 samples that have the highest "impact" score as defined in our Mention Field Definitions developer documentation. This is a standard sorting method for the display of mentions snippets used throughout the product.