I’m a procrastinator. I put presentations off until the day before. Like tomorrow’s session on AI and productivity - I’m prepping it tonight.
I like to tell myself I work better under pressure. It’s bullshit. No one works better under pressure. What actually happens: I either over-prepare and psych myself out, or I wing it and deliver something serviceable but not great. Lots of “ums” and “ahs.” Absent-minded professor vibes.
AI didn’t fix this. But it gave me a way to start before I’m ready. Brain dump the chaos, let the tool organize it, iterate through the mess until something coherent emerges. The process slowed me down, but it forced me to think with intention instead of winging it.
For work, I use Copilot exclusively - college-approved tool, proper data handling. For personal stuff, I bounce between ChatGPT and Claude depending on the task. I keep them separate because privacy matters. Don’t paste confidential data, student names, or work documents into web tools. Draft generic versions, get them cleaned up, then customize with the specifics.
Here are three things I’m doing with AI that actually work. Not the fancy stuff. The boring everyday tasks that save time.
1. Making Your Emails Actually Readable
This one’s humbling every single time I do it.
We write for ourselves, not our audience. Whether you’re sending emails to students with varying literacy levels, folks who use English as an additional language, or just busy people who don’t have time to decode paragraph three to find the deadline - we could all be clearer.
I have a bad habit: I use $10 words when $0.10 words would do. I like how language sounds as much as what it means. It’s persuasive, but it gets lost.
I already knew about bottom-line-up-front, inverted pyramid structure for long emails, all that stuff. But I still couldn’t write simply. So I started running my emails through ChatGPT. It was scathing. “Bring this down to accessible reading level.”
I’d used Hemingway app before - paste your text, clean up complicated sentences, fix heavy jargon. But it was always an extra step, an extra tool, an external webpage. I’d skip it. Using Copilot puts that clarity check right in the application where I’m already writing. That removed the friction. Now I do it.
Try this: Draft your email, then ask “rewrite this for grade 8 reading level while keeping all important information and deadlines. Make it clear and friendly.”
The result? Shorter subject lines. Bullet points instead of paragraphs. “Please make sure to” instead of “This correspondence is to inform you that.” Same information, half the words, twice as clear.
There’s nothing worse than putting stuff out there and not knowing what you wrote. I might as well be drunk.
2. Organizing Your Messy Thinking (AKA Reverse Prompting)
This one’s different, and it surprises people most. Most folks think AI is about asking a question and getting an answer. But you can also dump your scattered brain onto the screen and let AI help you think.
I do this with meeting notes all the time. I’ll paste in my messy notes and ask the AI to interrogate me for what’s missing. “What assumptions am I making here?” “What did you mean by X?” It asks better questions than I would’ve thought to ask myself. Then I fill in the gaps and suddenly the notes are useful.
Or if you’re a student with a group project mess - three people with different ideas, a presentation due Friday, nobody knows who’s doing what:
Prompt: “Group project on sustainable energy. Sarah wants to focus on solar, Mike says wind power, I’m thinking hydro. We need a 15-minute presentation, research paper, and demo. Due Friday. Help organize this into tasks.”
Result: AI organizes tasks by deliverable, assigns realistic timelines, suggests who handles what based on the interests you mentioned. You still negotiate with your group, but now you’ve got a starting point instead of chaos.
Optional follow-up: “Ask me smart one-at-a-time questions to help clarify what should be the focus.”
And it does. “What’s the most urgent concern?” “Are any of these required vs. optional?” You answer based on your context, and it refines the output.
One lesson I learned too slowly: I used to try having one giant endless chat dedicated to a topic. Story development in one thread, tool management in another, everything piling up for weeks. Bad idea. The context window drifts. The AI starts forgetting what happened earlier. I’d spend half my time re-reminding it what we’d already covered, what it was supposed to remember.
Eventually I learned: clear instructions, clear intention, fresh context. Don’t treat it like Google with keyword searches. Give it the goal - what you want and why it matters - then let it figure out the how.
3. Finding Current Information (Without Falling Into a Rabbit Hole)
My brain goes down rabbit holes. I’ll mean to catch up on one higher ed policy change, and three hours later I’m reading the 47th comment thread about implementation details I’ll never need.
AI helps me set boundaries I won’t set myself. You know how you mean to stay current with what’s happening in your field, but who has time to read trade publications every day? Web-grounded AI pulls recent articles and research for you.
Example prompt: “Find recent articles about [your specific topic] in [your specific context]”
Notice the specificity - community colleges not just “higher education,” administrative challenges not just “leadership.” The more specific you are, the less sorting you do afterward.
What you get: Recent articles (often from the past few months), themes that match your day-to-day reality, and you can ask it to “create a summary report” as a follow-up if you want the highlights organized.
Is this going to replace reading? No. But it’ll get you 80% of the way there when you need to know “what’s current on this topic” without spending an hour scrolling.
The Real Talk
I’ve tried seven AI tools this year. Six are gone. ChatGPT Plus? Too expensive in Canadian dollars. Mistral? Too niche. NotebookLM? Solved problems I don’t have. Perplexity Pro? Got a free year, we’ll see if it sticks. The graveyard is how you figure out what works.
What stuck: Tools that removed friction. Tools that let me start before I’m ready. Tools that make me think with intention instead of winging it.
None of this makes you an AI expert. The table stakes right now aren’t about building custom GPTs or understanding transformer models. They’re about writing clearer messages, organizing scattered thinking, and asking better questions.
Pick one of these three things. Try it this week. See if it saves you 15 minutes.
I’m still a procrastinator. The tools just force me to be a better one.
I’m just five minutes ahead of you on this stuff, and I’m still figuring it out. The tools will keep changing. But those five minutes have saved me hours.
