Tips for Appropriate and Effective Use of AI

Marketing Viewpoint by Ruth Winett

Does your company have AI guidelines?

Governments, industry, and educational institutions are struggling to develop AI policies and regulations. Some organizations have adopted AI policies that address key issues, such as the process (how the work is done), the output quality (the reliability of the output), security/policies, and the style and form of the output.

Following are useful AI guidelines from Professor Allison Stanger (Middlebury College), Microsoft (Safety Tips for using AI at Work), and Atlassian (Core Principles).

Process

"Never outsource the important parts," such as the "argument,’’ "memories," and "thoughts….Writing is thinking made visible: that sentence that resists you does so because the thought inside it is unfinished." (Stanger)
"Use the machine as a reader, not a writer." AI’s best use is "adversarial….it is a fast, tireless, un-flinching critic" of your writing, including whether your evidence is sound. Importantly, ask AI to present "opposing views" for consideration. (Stanger)
"Hand the machine the mechanical work…," including "formatting, finding repetition, smoothing transitions." AI is a skilled editor. (Stanger)
"Humans lead all writing and ideation. AI can help validate, edit, and refine, but a human should always write the first and final drafts in their own voice. (Atlassian)

Output Quality

"Treat every factual claim AI conveys as suspicious." (Stanger)
Double-check AI-generated results. (Microsoft)
"Fact check, fact check, fact check" all AI output "before implementing—including any factual information, data, or quotes." (Atlassian)
"Guard against AI erasure. AI defaults to the audiences most represented in its training data….When reviewing AI writing suggestions, pay as much attention to what it removes as what it adds." (Atlassian)
"Everything we publish needs to have a human byline." The author is "publicly accountable..."including anything AI-influenced." (Atlassian)
"Don’t publish anything you wouldn’t read yourself." (Atlassian)

Company Security/Government Regulations

Follow your organization’s policies for using AI at work. (Microsoft) Only use "company-authorized AI services." (Microsoft)
Don’t disclose "sensitive or confidential company details" to an AI service, including "personal information about employees or vendors." (Microsoft)

Style/Form

"Know your own voice and defend it." (Stanger).
"When using AI to edit or refine, always specify [in the prompt] the text’s audience, purpose, and tone to avoid generic feedback." (Atlassian)

AI Guidelines Are Evolving

AI guidelines are essential. Guidelines tell employees when and how to use AI effectively. Guidelines also tell employees how to secure company data and AI output. However, guidelines will differ for each department. For example, finance and marketing departments have distinct functions, use different sources, create different outputs, and require different policies. Ultimately, the output should resemble the work of a human, not a machine and an algorithm.

As some AI services have hacked into other systems autonomously, governments must also develop – and enforce- sound AI policies and regulations.

Sources:

"AI Makes Me a Better Writer," Allison Stranger, Middlebury College Professor, Wall Street Journal, 8/18/26.

https://www.atlassian.com/blog/ai-at-work/atlassians-guidelines-for-writing-with-ai [See the discussion on when to use and when not to use AI.]

Safety tips for using AI at work. Microsoft Support.

Acceptable Use Policy for generative AI Google Cloud

Note: I asked ChatGPT to list companies that have AI guidelines and name sources used. Then I consulted the source documents.

Market Insights

For Email Marketing you can trust

articles index | home | back