AI at Work
AI can be a powerful tool for refining everyday work, but organizations must protect information and keep humans in the loop
There are a lot of questions about artificial intelligence in the workplace, but not a lot of answers when it comes to when and how to use it best. Navigating carefully to protect proprietary information is critical, but there are some much lower-risk approaches that yield valuable efficiencies. While I don’t think anyone should see AI as the first and last stop on any project, it can be a good “refiner” of content you have already created yourself.
Protect your intellectual property
At FGIA, we have carefully deliberated the roles AI should and should not play in carrying out daily work. When making these decisions, we had to think about AI’s limitations and risks. For example, a big one is the threat to intellectual property. Once something is uploaded to AI, its ownership and accessibility to others becomes precarious, especially without using a paid version that may offer additional security.
With respect to privacy concerns and the protection of confidential and proprietary information, FGIA does not use AI for document development or for minute-taking for committee and task group meetings. But in more appropriate situations, AI can be a very useful tool. Try to think about using AI as an “amplifier” rather than as a “creator.” Through that lens, non-proprietary content can be uploaded for refinement, which can greatly expedite the completion of daily work and create much-needed efficiencies, given how many employees are stretched thinner than ever.
Save time with AI
For our part, FGIA has identified the following as time-saving examples of AI use, some of which you may be able to apply at your own company:
-
Polishing or abbreviating content, like reports, or checking for grammar
-
Revising the tone of content to make it more conversational or more formal
-
Refining promotional messages and standard customer communications
-
Creating renderings for stage or room layouts for events
-
Developing initial, not final, draft agreements to reduce legal fees
-
Producing an annual report of accomplishments and value proposition
-
Tracking regulatory and legislative issues to learn about new developments
-
Notetaking for internal or vendor meetings (without confidential or proprietary information) if all parties involved agree to its use
Content review
The above types of AI inputs are lower risk because the prompts are more innocuous since they don’t include confidential or proprietary information. However, it is critical for all AI-generated content to be thoroughly reviewed by a human for accuracy, tone and context before being shared externally. This ensures quality control and prevents the spread of misinformation or off-brand messaging.
Copyright
Another interesting part of AI is the question of copyright. Anything generated solely by AI may not be copyrightable by an organization. Human creative input is necessary to claim ownership. It is generally recommended that AI users keep total AI content under 20 to 30% of a final product. Some additional important tips include knowing the parameters of your AI tool’s terms of service and complying with your company’s policy on using AI.
It may feel like the wild, wild west now, but AI certainly has potential. Just remember that the human factor is essential to any final work.