This feature is currently in an open beta and opt-out only. Visit the section below for instructions on how to opt-out of AI features.
In this article:
About Iris writing assistant (powered by OpenAI GPT)
Iris is Brandwatch’s AI. It uses cutting-edge proprietary and integrated AI models to help you work faster and smarter. The Iris writing assistant leverages OpenAI GPT (Generative Pre-trained Transformer) technology to offer assistance when writing and editing your text in Publish, Engage, and Advertise.
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 and generate human-like text. GPT models excel at tasks like language translation, text completion, and even generating creative content. Visit Open AI’s website to learn more about OpenAI and ChatGPT.
The Iris writing assistant can support different user roles in their day-to-day tasks in the platform:
- Support agents and community managers using Engage can improve their replies and communications, for both public conversations and direct messages.
- Content creators and editors utilizing Publish can enhance their social media posts.
- Advertisers working on their ad copy in Advertise can optimize their messaging.
When composing text in Engage, Advertise, or Publish’s text editor, you can request assistance from Iris to optimize or edit your text. Iris will offer a suggestion and you can then choose to accept, reject, or request a new suggestion based on your needs. If you accept a suggestion from Iris, you can always further edit the text if needed. The writing assistant feature is offered across all supported social networks.
In Publish, you can use Iris’s Generate post option to create effective and intentional content for your brand. Choose a tone of voice or brand personality for your generated posts, include emojis if desired, and specify a text length for your posts. Iris will provide you several generated posts to choose from or modify.
In Engage, you can use the Generate reply feature for Iris to automatically generate a reply message in an Engage thread based on the conversation context.
Iris also offers the following editing capabilities across Publish, Engage, and Advertise:
- Grammar and spelling correction: Identifies and fixes mistakes in your text.
- Text rephrasing: Suggests improvements to enhance the clarity and effectiveness of your writing.
- Text length modification: Makes your text shorter or longer based on your requirements.
- Synonym suggestions: Offers suggestions for alternative words.
- Emoji suggestions: Recommends suitable emojis to enhance the emotional tone or engagement of your text.
Using the writing assistant
When using the text editor in Publish, Engage, or Advertise, you can access the Iris writing assistant by clicking on the wand icon in the lower-right corner of the text box. From there, you can select the assistive feature you would like to use from the drop-down menu.
Generating posts in Publish
When using Iris in Publish, select the Generate post option to create content.
Enter some starter text in the text box for your post. The more detailed your text, the more detailed Iris’s generated posts will be.
Under the Preset tab, you can select one of 12 brand personalities to apply a personality to your post. Under the Custom tab, you can create your own personality by combining different parameter options under formality, emotionality, distance to the reader, and humor.
After you choose personality options under either tab, you can select a Post intention to further define the language and content for your post (e.g. Advertising, Announcement, Educational, etc.), specify the post length, and select whether or not to include emojis.
Post length presents are calculated based on the historical data in Brandwatch for each network.
When you are finished selecting your content preferences, click Generate content at the bottom of the window. Iris will take a few seconds to generate the content, and then three post options will populate in the window.
If desired, you can further adjust and refine the text and options you input into the tool and then click Generate more for Iris to generate additional post options in the window. While the window is open, Iris will keep the history of suggestions so you can compare and/or use the previous suggestions.
Click Apply to draft under a generated post option to insert that post text into your Publish post.
Generating a reply in Engage
In Engage, you can use the Generate reply option in Iris to create a reply message based on the conversation context. Click the wand icon in the reply text box in Engage to use Iris. The Generate reply option will be available as long as you have not entered any text in the reply text box yet.
After a moment, Iris will populate a reply message. You can choose to reject the generated message, refresh it to generate a new option, or accept it to add it to the reply text box in Engage. You have the option to further edit or customize the reply message before sending it.
If you would like to provide Iris some starter text or instructions to include in your reply message rather than generating a reply based solely on the conversation context, enter your starter text into the reply text box and then use Iris's Make longer option to generate a longer reply based on the text you have provided.
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.
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.
Is my data private when using the Iris writing assistant?
We prioritize the privacy and security of user data. The text processed by the writing assistant is treated as confidential and is not stored or used for any purpose other than generating suggestions. ChatGPT doesn’t use the submitted text for training purposes.
Are there any rate limits for using the Iris writing assistant features?
To ensure smooth performance and prevent abuse, there are rate limits in place for using the writing assistant features. This means that there is a cap on how often someone can repeat an action within a certain timeframe. These limits are set to maintain a high-quality user experience while also preserving system resources. If you encounter any rate limit notifications or have concerns about specific usage scenarios, please reach out to our support team for further assistance.
Are there limitations to how many times I can use the Iris writing assistant features?
Besides the rate limit explained above, there are no limits on how often you can use the features. You can use the features multiple times with a post or reply, and for as many posts and replies you need.
Are there any limitations on the length of the text that I can submit for suggestions to the Iris writing assistant features?
Yes — limitations are imposed by OpenAI on the data processed by the features. Roughly, up to 2,000 characters can be processed, which includes both the text written by the users and the suggestion returned by the feature. There are also limitations on the amount of requests we can send to OpenAI per minute.
Will these features work for different languages?
Iris Writing Assistant works in the languages that the user uses while writing. It supports all languages that ChatGPT supports.
Currently, the Iris Writing Assistant is highly proficient in understanding and generating English text. The underlying ChatGPT models were trained on a large dataset of text from the internet, and as a result, it can generate text in a wide variety of languages, including English, Spanish, French, Chinese, Russian, and many others (95 languages in total). However, it is essential to note that the model's performance will vary depending on the language and the complexity of the text being generated. Therefore, the model’s performance and accuracy might be higher for English compared to other languages.
Can I customize the suggestions provided by the writing assistant?
Yes, after accepting the suggestion, the user can always edit the text, as if they were writing the text in the editor by themselves.