We asked 20+ product managers how they use AI. Here’s what we learned.

If you’re a product manager, you’ve seen your fair share of AI tools aimed at PMs: Use AI to create PRDs, use AI to write tickets, use AI to replace yourself!

But there’s a ton of hype in AI: Like any tech trend, we’ll look back in a few years and see some tools as game changers while we remember others as once-intriguing ideas.

I wanted to know not which AI tools are being built for product managers, but which AI tools product managers use to make their day-to-day easier. So I asked a bunch of product managers in communities like Reforge, Lenny's and Reddit! Here’s what I found out.

Spoiler: It wasn’t this (I wish it was though)

How Miroslav Pavelek thinks about AI

Before we get into any other responses, we wanted to feature one PM in depth. While lists and short quotes are useful, it can be useful to understand a deeper perspective.

That’s why I interviewed Miroslav Pavelek, who’s a product & project manager at Vaccumlabs. Here’s his take on AI:

What AI tools/ChatGPT use cases do you use in your day-to-day?

ChatGPT:

• Whenever I write any report (e.g. to stakeholders), I let ChatGPT polish it as I am not a native speaker)
• I use it as an advisor - e.g. if I am unsure about how to react in some situation, I ask ChatGPT for opinion as PM expert, so I gain different perspective on the issue. It can also roleplay as difficult client/customer, so I can gain some initial insights without real talking to people.
• Product requirements- I write 2 bullet points specification and desired structure (e.g. divide it into 5 user stories, write 5 acceptance criterias for each US). It serves as nice initial draft, as it can make 70 % of the writing. Of course, I have to rewrite it and add my points to make it to my wish, but it definitely speeds my work a lot.

How has AI impacted your work?

I can do typical "boring" work much faster and I can get get second opinion without any hassle. So, my efficiency improved a lot.

What process you wish there was an AI tool for?

Making Wireframes out of PRDs/User Stories - I would appreciate having some simple diagrams based on PRD or Users Stories, which could be further polished by chat-like interface.

What product task do you think will always need to be done by humans?

Real talking to (potential) users - human always prefers talking to another humans, so gaining qualitative insights will stay regardless of AI. Also setting up product vision, as it must be done by product manager/founder of company.

Miroslav’s perspective is super interesting and I loved the specificity he provided for his ChatGPT use cases. One thing’s for certain: He won’t be one of the PMs left behind by AI. Now, I also gathered briefer responses in product management communities. Here are the insights I gathered:

80% of it was “just” ChatGPT

When I asked about AI tools in product management communities, I expected hidden gems—secret tools just out of beta that would transform how I do my daily work. That didn’t happen. While there were some (and we’ll get to those), most PMs simply use ChatGPT for different use cases:

PMs use ChatGPT for writing. A lot.

This was interesting: Product managers aren’t writers (at least that’s not the job they signed up for). And it can feel like writing all these documents is an obstacle to what you love: Talking to users and envisioning products people will love.

That’s why it’s not surprising that responses like the following trickled in:

• “Chat GPT to not make my emails sound like a passive aggressive PM” (Reddit)
• “ChatGPT to get drafts started. Literally everything that needs writing. From PRDs to internal posts etc. it’s just a tool that I use so that I don’t have to start from a blank page every time.” (Reddit)
• “It's mostly been useful for me with copywriting. I'm not the best writer so after I jot down everything I have to say, I will just run it in chatgpt and ask to simplify my draft.” (Reddit)
• “I use a tool to break down work that I like, and chatgpt to create some documentation and slack comms” (Reddit)
• “Whenever I write any report (e.g. to stakeholders), I let ChatGPT polish it as I am not a native speaker)” (Lenny’s Newsletter Slack)

I could keep citing examples, but there’s a theme around automating/polishing your writing.

But let’s be real—we all really want this AI writing tool:

ChatGPT to understand technicalities

Rifts between engineers and PMs often happen because engineers think the product manager doesn’t really understand what they do—often because they lack technical understanding.

ChatGPT is surprisingly good at understanding and writing code, which could power-up non-technical product managers. Some on Reddit seem to find that useful:

• “My employer uses Google office suite, so I’ve been using their AI tool Bard a bit, it’s been useful in the respect of asking technical questions or provide insights/framework on how to approach certain problems/situations.”
• “ChatGPT/Bard/Claude to better understand technicalities, to tweak/fix issues with SQL queries/Excel formulas and to debug API calls for external integrations”
• “Plain ChatGPT helps you to understand technicalities in case you want to test your own software until it breaks. That’s the only AI actually helping me currently.”
• “I use it for quickly throwing together apps scripts and automating basic tasks. It can knock out a simple 40 line script faster than I can possibly type it, with no errors.”

ChatGPT for PM coaching

Quite a few PMs replied that they used ChatGPT to chat through situations and use it as an advisor/coach:

• “ChatGPT, especially the new Vision and chat features, chatting to the AI on lunchtime walks is just a remarkable way of thinking through a problem.”
• “ I use it as an advisor - e.g. if I am unsure about how to react in some situation, I ask ChatGPT for opinion as PM expert, so I gain different perspective”

As mentioned, ChatGPT seems to be the tool of choice for a lot of product managers. But there are specialized tools that were mentioned too:

Specialized AI tools for product managers

• Scribe was mentioned multiple times to automate documentation. Scribe sits in your browser and records what you do before transcribing your actions with an LLM. This automates writing the documentation!
• AI by Zapier was mentioned to automate some workflows in Jira.
• Stable diffusion and Midjourney were mentioned to create visuals for PowerPoints… and for memes for work group chats (important too!)
• Copy.ai was mentioned for, well, writing first drafts of copy.
• Prezent was mentioned to build great business presentations faster with AI
• Zoom’s AI assistant was mentioned to automate taking meeting notes.

Wrapping Up: PMs love AI tools

Product managers usually work in tech and love trying out new products. That puts them at the forefront of new B2B AI tools.

While some of the specialized tools were interesting, I was more surprised by the various use cases for ChatGPT—showing once again what a behemoth of a product OpenAI has created.

With there being such a clear winner (for now), it’s exciting to see which specialized tools emerge over the coming years. While it feels like the entire tech ecosystem changed shape when ChatGPT launched in November ‘22, it’s easy to forget it’s only been one year.

If the past is any guide to the future, the next few years will surprise us (both positively and negatively) for how we product people do our work in the future.