In a fresh development, instead, they always hit me when I’m doing something else: taking a walk, commuting to work, or other “dead times” when my brain is free to wander. Small-but-important tasks never come to me when I’m actually typing my daily to-do list.
In a fresh development, that’s often when your best ideas come out, particularly when pertaining to daily tasks. But who wants to plow through all those disjointed ramblings? One habit of truly organized people (I’m not one of them, sadly) is to fire up a voice memo app and just talk.
The report highlights that or any other respectable AI chatbot, like Claude or Gemini. They all excel at taking large data dumps—anything from lengthy error logs to random stuff you said while strolling in the park—and finding order within the chaos. Well, ChatGPT, that’s who.
According to the latest update, but AI voice modes tend to be overly chatty, interrupting your trains of thought with unwanted praise (“You hit upon something profound, Ben!”) or unnecessary follow-up questions (“Would you like me to dig up reviews for that plumber?”). Now, you could simply activate the voice modes of ChatGPT, Claude, or Gemini and talk directly to them about your daily to-dos.
According to the latest update, you’ll also need a prompt that a) turns your ramblings into an organized to-do list and b) extracts your thoughts without adding anything else. Instead, I recommend using a traditional voice memo app—Apple Notes will do, or Google Recorder for Android users—and then handing a transcription of your musings to your preferred AI chatbot.
According to the latest update, the beauty of this method is that it works with any AI (not just those with voice modes) and it also works fine with free AI accounts.
As part of the ongoing story, here’s the prompt, culled from a variety of versions on the web:.
As part of the ongoing story, don’t clean it up or rewrite it. Instead, just extract:. Here’s a rough voice note I just dictated—treat it as messy, stream-of-consciousness input.
Industry observers note that every concrete task or next step I mentioned. 1.
In a fresh development, any idea I said I wanted to revisit or think about later. 3.
In a fresh development, don’t add, suggest, or infer anything fresh. Only pull out what I actually said.
The report highlights that naturally, top-of-mind stuff came first (“Um, first, I need to write the prompt of the week”), but other random to-dos popped into my head as well (“Oh, make sure to get a pepper refill because we’re running out of, uh, pepper”). I tried the prompt using Apple Notes and all I did was talk out loud, going over every possible task I could think of.
The report highlights that in less than a minute, my to-do list arrived:. I then took the raw transcription and fed it to ChatGPT (using the GPT-5.5 Instant model).
Industry observers note that here’s what you actually mentioned, organized by category.
As part of the ongoing story, — Write the “prompt of the week.”— Come up with a topic for the weekly newsletter.— Call the dermatologist about paying a bill.— Do a quick check of your bank accounts.— Look at your list of weekly dinners to decide what to make for dinner tonight.— Charge your Sonicare toothbrush.— Start packing for your trip to Recent Jersey this weekend.— Get a pepper refill because you’re running low.— Clean the tanks for the robot vacuum cleaner.— Finish the laundry.— Look over the issue with BenBot and see if it needs to be followed up.
As part of the ongoing story, — Decide what to make for dinner tonight.— Decide whether the BenBot issue needs follow-up.
In a fresh development, not bad, although I’d say picking a newsletter topic belongs in the “decisions you need to make” section (careful prompting could fix that).
According to the latest update, for extra credit, you could also use a connector to send your to-dos to Notion or another online workspace. Give it a try, and don’t be afraid to tweak the prompt for free-flowing brainstorm sessions.
As part of the ongoing story, his coverage of artificial intelligence interrogates the most recent LLMs, and how they can be used at work and at home to be best prepared for the AI revolution. “AI is going to change our lives sooner than we think,” Ben writes. “Our best way to adapt is by using it every day.” Ben has been a PCWorld author since 2014, and has covered everything from laptops to security cameras before launching PCWorld’s AI beat. Ben's articles have also appeared in PC Magazine, TIME, Wired, CNET, Men's Fitness, Mobile Magazine, and more. Ben holds a master's degree in English literature. Ben has been writing about consumer technology for more than 20 years, and now focuses his reporting on AI as it relates to the basic human experience.