Share
With AI skills being highly sought after in the current job market, it’s clear that companies want their employees to know how to use AI. Your company will likely have a set of general policies about its expectations when it comes to where AI fits into the workflow, but it can’t account for every situation. Knowing when you should and shouldn’t use AI in your work is an important judgment skill to have. If you’re unsure of where to draw the line with AI use in the workplace, here’s what you need to know.
When to Use AI
Good times to use AI resources are when the work is more low-stakes and nobody is going to get fired if the output is wrong. These are also situations where there isn’t any sensitive information involved, you can check the output and make the final decision, and AI is helping you think instead of thinking for you. There are plenty of scenarios in the workplace that are going to fit this bill.
Communication
AI is a great tool to use for communication purposes. Some examples of ways to use AI to help you communicate better are by drafting simple emails, changing the tone of an email, checking for grammar mistakes, and turning notes into a full paragraph. All of these situations allow for clear communication, and you are still able to tweak the output if AI makes a choice that you don’t agree with.
Learning
Using AI to learn more is also a great idea, especially when you’re trying to get your bearings in your first corporate role. Some instances where AI use is okay are understanding corporate jargon, figuring out the best Excel formula to use, summarizing current news, and breaking down a complex topic so you can better understand it. These situations will help you learn and be successful in your role.
Productivity
The benefits of AI are usually tied to efficiency. Using AI to make yourself more efficient and productive can help you be more successful at work. AI can help you create your to-do list, turn meeting notes into action items, create meeting agendas, and create a process checklist for a routine task. There’s no harm in using AI for these tasks because they are low-risk, and if you don’t agree with what it creates, you have the opportunity to do something different.
When to Avoid AI
While AI can improve efficiency, it shouldn’t be used for everything you do at work. There are a lot of situations in which using these tools can result in leaked data, inaccurate results being presented to clients, and data being incorrect due to AI making a mistake. Here are some instances where you should avoid using AI.
Confidential Information
Inputting confidential or sensitive information into AI tools should be avoided. Information given to these tools can be saved in the platform and used for training purposes, making the information available to a third party. There have been reported cases of these instances causing major issues for companies. Back in 2023, Samsung restricted the use of generative AI for its workforce after employees gave confidential information to ChatGPT. This meant that OpenAI had sensitive information about the company, with no clear way to get it deleted. Nowadays there are some guardrails in place, such as AI tools offering enterprise accounts that restrict user data from being used for training purposes. But even with these protections in place, it’s still best practice to avoid giving AI tools any confidential data when you aren’t explicitly approved to do so by your company. You don’t want to be the next Samsung story.
High-Stakes Accuracy
As great as AI tools can be for companies, sometimes the tools get it wrong. These tools have been shown to make mistakes and sometimes hallucinate—i.e., giving false or made-up information. If you are working on a high-stakes project that needs 100% accuracy, don’t rely entirely on AI. Check final numbers, reports, or client deliverables to ensure accuracy. Giving a client false information is something that can have very serious consequences, so AI isn’t for work that you need to be positive is perfect.
Personal Conversations
Some communications should be entirely human and not AI-assisted. If a friend or coworker is discussing something personal with you, the last thing they want is a response that AI wrote. They want to hear from a real human and want to know what you really think, not what ChatGPT thinks—otherwise they would have told a chatbot directly. If a coworker confides in you about something deeply important to them, it would be incredibly inconsiderate to give them a generic AI response. Sometimes people just need to talk to other people.
At the end of the day, use your best judgment on whether or not a situation calls for AI use or AI avoidance. If you are ever unsure, check your company's AI policies to see what you need to avoid.
Share
