AI-Driven Tools

Overview
Businesses are adopting AI technologies for "business model innovation, process transformation, disruption and gaining competitive advantage." In some cases, AI technologies will replace human labor, but much of these adoptions are intended to add to and improve human intelligence. This AI-human partnership has been referred to as collaborative intelligence. While technical communicators (TCs) already use widespread tools such as spelling and grammar checks as well as predictive text, they can use more complex AI technologies that can boost their efficiency and expand their capacity to meet the information needs of their target users.

Tools for User Advocacy
Numerous sources refer to the important role that data analytics, propelled by AI, will intersect with TC work. . AI-driven tools are used for data collection and analytics that allow organizations to identify and respond to client needs and improve operations. Collectively, such data collection, analysis, and decision-making, when facilitated by technology, is referred to as business intelligence (BI). For example, AI can facilitate the analysis of users' interactions with a company's product or service. TCs will use BI to develop or revise documentation or training material that is better suited to users.

TCs may also be involved in contextualizing this data for other stakeholders through storytelling or visualization. Given their skills at rhetorical analysis, TCs are in a good position to help identify which data is actionable. AI systems can also, in turn, use these analytics and reactively send information back to users. TCs may be involved as a human collaborator who oversees the AI to ensure appropriate interventions are deployed.

AI tools will deliver biased information if biases are not addressed when they are first developed. Scholars have underscored the importance of training TCs to understand how AI may exclude or misrepresent users and thus produce suspect data. They recommend that TCs have an understanding of data science principles. This allows TCs to be a user advocate who can identify when AI is underused, misused, or overused.

Tools for Data Reporting & Visualization
Two tools that facilitate the organization and visualization of data are Google Cloud's Looker, Google Looker Studio, and Salesforce's Tableau.

Google Cloud's Looker is a BI platform that incorporates AI. It acts as a single source of an organization's data which can produce accurate and consistent queries across the enterprise. Data analysts, product managers and developers can use it to create custom applications and embed analytics into a TC's workflow. Looker's programming allows TCs to develop easy-to-read reports and data dashboards with visualizations that make pattern analysis more accessible.

Google's Looker Studio is a free, self-service application that allows users to create data visualizations, dashboards, and reports based on the user's own data. See figure 1. Tableau is a competitor to Google's Looker and is owned by Salesforce. It is also a BI platform that leverages AI to generate rapid and useful data visualizations that allow users to identify and act on patterns in the data. Tableau also has a free, public platform that gives users access to publicly available data. Users can use the platform to learn about and practice data visualization skills. See figure 2.

Tools for Text Analytics
AI can be programmed to analyze texts for several purposes useful to TCs. Text analytics are AI-driven processes that can examine a large quantity of qualitative, textual data (often referred to as unstructured by data scientists) [13]. These processes can:


 * Provide statistical information such as frequency and prevalence of a given term or phrase
 * Categorize and package textual information
 * Route packaged information to those responsible for responding to it
 * Detect plagiarism
 * Assign a readability score to a passage
 * Assign a probability score that a given text was produced or paraphrased by an AI content generator

Tools for Sentiment Analysis
Sentiment analysis is a tool that can be used for audience analysis. Here, the AI tool examines textual data for words and phrases that signal an emotional response and assign a sentiment score on a scale from negative to positive. See figure 3.

Sentiment analysis has also expanded to data gathered from wearable devices such as fitness trackers that monitor heart and respiration rates, blood pressure, movement, and mood.

The TC can use data from sentiment analysis tools to gauge user needs prior to developing documentation. They can also use such tools to analyze user responses to the design and delivery of information.

Researchers have raised concerns related to the ethical use of sentiment analysis tools, given that they are instruments of surveillance capitalism. Professor Shoshana Zuboff used this term to describe how businesses can use sentiment analysis to predict and shape behavior for profit or political gain. Here, TCs may need to understand how their work is connected to this business practice.

Tools for Content Generation
Open AI's third generation Generative Pre-trained Transformer (GPT-3) language generator is an AI technology capable of producing text that closely imitates human language. A user gives the system a prompt and some introductory text, and it can generate understandable passages of varying lengths. It has been used to generate articles such as sport and business reports that follow a regular format. It is also used to spark ideas for creative writers.

According to Robert Dale of the Language Technology Group, GPT-3 may be useful for generating content for creative work, but cannot be trusted to provide accurate, technical information consistently. It requires a human editor and fact-checker. However, it can be useful for workplace communications. For example, OtherSideAI's HyperWrite, an application built from GPT-3, can produce complete emails that imitate the author's writing style if given an outline of key points.

The audience for many of the current GPT-3 based applications are content creators and copywriters who work in marketing. These applications include Jasper.ai, Copy.ai, and Rytr.me. Commentators from this arena express similar concerns about the usefulness of the technology. They note that while it helps with idea generation and search engine optimization tasks, it does not have the creativity and empathy that produce effective marketing content.

Tools for Workplace Productivity
Virtual assistants are AI applications that a TC may use to increase their workplace productivity. Apple's Siri, Microsoft's Cortana, and Amazon's Alexa are examples of virtual assistants that are used by the general public in the United States. Virtual assistants can be programmed to handle tasks more tailored to the workplace. For example, chatbots can be embedded into content management systems to fetch needed materials or analyze team members' schedules and identify meeting times.