Comment by haswell | Hacker News Reader

Comment by haswell | Hacker News Reader

Having worked quite deeply with Jira and a host of other enterprise products that target this set of use cases, and having a pretty good understanding of what decision makers are looking for when evaluating these products for fit (I have been that person), what does "AI First" mean?

I realize AI is the hype right now and it seems like most products have to pay the AI Tax (i.e. "Yes we're doing AI stuff so you should <buy/sell/fund/etc> our stuff"). But "AI" is a nebulous category, and when a product bills itself as "<Noun>-first", it sets up an expectation that <Noun> is fundamental to the product, i.e. if you stripped all other aspects of the product away, what's left at the end is <Noun>. But I have no idea what that means in the context of the current "AI" hype.

Based on your demos, you've built a fairly standard looking ticket tracking tool that has some AI features. And those are features that every incumbent in this space started adding to their products years ago or are actively doing it now.

I mention this not because I'm trying to say your product doesn't have value, but because the way you're positioning this doesn't make a lot of sense to me. As a prospective buyer, if I'm already using an existing tool, the moment I start digging deeper to know what "AI First" means, I'll find that what this really means is "Jira but with some AI features on top", which isn't very compelling when my <Vendor> account team has been telling me all about their new AI features that I can start using on all of my existing data as soon as I upgrade, no 6-18 month migration required.

If I'm a frustrated customer of those products, AI is not the reason I'll be looking to switch, and I'd be far more likely to be interested in performance, extensibility, openness, integration capabilities, etc.

Maybe that's not the type of customer you intend to target, but if you hope to reach them, I leave this as food for thought. Best of luck to you.

Someone had to call it out!!! thank you for that.

Oh, the AI hype! It's basically just common sense wrapped in a shiny new package.

Let's call quick filters 'AI' and be amazed, shall we?

When the dataset is limited, what magical tricks can AI pull? Joining threads or conversations is just a fancy wildcard search. Where's the real optimization?

all that i wanted to say, only better.

i am a zero-to-one guy and i get that this kind of positioning is often forward-looking, but i think that's what early-stage founders get wrong, especially if they aren't core whatever-noun they are anchoring to.

if you consider all other incumbents in this space, they are all AI-driven. your differentiator could be AI only if you are core AI and have drawn up models for project or work management use cases. otherwise, just stick to whatever your differentiators are in the near-term and change positioning later when you are core AI.

Taskade did this well. Notion is doing it well. they are all pivoting around AI without trading off whatever's their core.

> If I'm a frustrated customer of those products, AI is not the reason I'll be looking to switch, and I'd be far more likely to be interested in performance, extensibility, openness, integration capabilities, etc.

Sounds like this is what they should be asking you more about.

Would you mind rattling off some a wish list?

Perf I can imagine but some of the others not so much in detail.

> as soon as I upgrade, no 6-18 month migration required

Maybe not a 6-18 month migration, but you will have a similar wait for Atlassian to deliver something approaching functional.

Of course this doesn’t matter to the leaders that can now claim their ticket tracking is AI enhanced, but the net value of the feature at that time is negative or zero.

Otherwise I agree though. Frustrated Atlassian customers do not move because they really wanted AI.

Really appreciate the thoughts. The places where we believe AI matters a lot is

1. Ensuring ticket has enough information 2. Offloading some amount task before someone picks it.

These are the 2 key areas we are focusing and we will launch some features around these in the coming weeks. Would love to hear if there are some other places you feel we can focus on?

What about, instead of using AI to summarize the issue text for the developer, using AI on the _submission_ end to notice if the bug report doesn't contain enough details and ask the submitter clarifying questions until the submission is a really good bug report?

> what does "AI First" mean?

Copious amounts of gratuitous hallucinations contaminating the data you are looking to work with.

Reply on news.ycombinator.com