By Teresa Torres · producttalk.org · @ttorres on X · LinkedIn
Teresa Torres and Petra Wille argue that people who feel their problems are too small for AI are mistaken; the real blocker is lack of exposure to what AI is good at. Teresa describes her habit of testing one to-do item with AI each day, which moved her from feeling behind to using it confidently. They stress that starting with poor results is part of learning, that the tool hype (MCP servers, plugins, new apps) is noise, and that community and simple, task-focused experimentation accelerate progress. The episode is a nudge to begin with one small task for about 15 minutes.
01Key takeaways
- If you have a to-do list, you already have problems AI can help with.
- Pick one small, concrete task and test it with AI each day.
- Expect the first attempts to be bad; iteration is where the learning happens.
- Ignore tool hype like MCP servers and plugins until a specific need appears.
- Join communities or show-and-tell sessions to learn faster than from tools alone.
- Start with a simple prompt such as asking AI how it can help with a task you must do.
02Key sections
- Small problems are still valid
- The hosts open by challenging the belief that everyday problems are too minor for AI. They frame the obstacle as a lack of exposure rather than a lack of suitable problems.
- A daily habit of testing one task
- Teresa explains that trying one to-do item with AI every day built her confidence and fluency. Consistent small experiments matter more than occasional big ones.
- Embrace bad first results
- Early outputs are expected to be poor, and iterating from weak results is how people learn what AI does well. Petra's image-generation story illustrates moving from terrible to good.
- Filter out the noise and find community
- Tools like MCP servers and new apps are not prerequisites for real value. Show-and-tell sessions, blogs, videos and local meetups speed learning more than any tool.
- A simple way to start
- The hosts recommend a low-stakes first prompt, asking AI directly how it can help with a task you must do, and spending about 15 minutes playing.
03From the post
“Listen to this episode on: Spotify | Apple Podcasts Think you don't have big enough problems for AI to help with? Think again. In this episode, Petra and Teresa tackle one of the most common blockers people face when starting out with AI: not knowing where to begin. It's”
Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.