One of the worst ideas I've heard in a while. A company with the premier LLM, asking companies to outsource the platform running all internal communications. What does OP think we are all doing here in business? This is the Ycombinator community edition of Rodney King's famous "Why can't we all just get along".

> Microsoft did, and Teams is by all reports a solid success.

Not sure if the author has used Teams.

But otherwise, I agree we need an actual good, adorable Slack clone. I thought Google might do this after not buying Slack, but I'm not hearing anything about their solution.

I can't read the article, but I feel people are missing the point here.

Slack is a really really good product because it is simple enough and works nice - performant, has just enough features but not too many and the UX/UI is good.

Its not a power tool but it gets the job done without getting in the way. You would know what I mean if you have used teams/ google chat etc.

Sure you can criticise slack for being a bit slow, not having nested threads.

For context: slack is the main app I use at work and spend a * lot * of time there.

But OpenAI _can_ beat Slack at these things if they have the technical acumen. But real differentiator comes in having an all in one platform that can help you run workflows. Recall that ChatGPT UI is fundamentally a chat box. If ChatGPT can integrate common workflows like

- send an email to a colleague for something

- schedule a meeting at a certain time

- deploy to production

- approve leaves

- create quick code changes with natural language like "change threshold to 50 in my repo"

- integration with observability and alerting

Then you don't have to leave this tool at all. There's a lot of potential here.

I'd like to speculate, with the recent success of AI agents on the command line with OpenClaw, that perhaps IRC could be the future of AI-enabled chat rooms?

> OpenAI spends time and money building a slack competitor, because they've apparently run out of good ideas

> Slack uses AI to improve the existing product

> Slack is still marginally better, so businesses continue paying for it

> OpenAI now on the hook for maintaining one of many cheap slack clones

> Investors are left scratching their heads...

Late stage bubble behavior

There are already a ton of slack alternatives. Slack connect is the main thing that is blocking a lot of people from moving off slack, otherwise chat is a commodity.

I use Slack every day, and I love it. Integrations are simple and reliable, giving us useful information about critical things.

Why it uses 400mb I have no idea.

Does anyone use Mattermost? I remember thinking it wasn't too bad, and I guess it's open source.

Funny, didn’t even mention using the massive amount of compute available to them to build it!

A prompt ran through a Wiggum loop over the course of a week/month and viola

> Slack has been on a slow rachet up in prices and has struggled to introduce compelling new AI features

I can think of a few reasons that Slack could be improved upon. But a lack of AI features is not on that list. Slack is effective for async communication between humans. We don't need AI features to accomplish that, and most AI would just be annoying slop. If you are using Slack for something else, maybe AI features would help those other uses, but you also might be stretching the cases for which Slack is a good thing.

We use Slack at work, and everyone we work with uses Slack, and we all work together with Slack Connect. I suspect if we moved to a competitor that’s pretty much the main impact we’d see, and it wouldn’t be good unless everyone else work with moved too. I think that network effect is probably the only meaningful differentiation in that space.

I'd rather it build docs. Or at least have a feature in chatgpt that lets you highlight something and start a comment thread, rather than a multi-page essay response as a continuation of the chat itself.

The real issue isn't whether OpenAI could build a Slack competitor — it's whether they should fragment their focus even further. They're already stretching into search, image gen, video, agents, and an app store. Every great platform company eventually gets the itch to become everything, and that's usually when quality starts slipping on the core product.

Why not ask for a federated slack?

What evidence is there that OpenAI will be more benevolent than Salesforce? Perhaps we shouldn’t give large corporations more opportunities for data mining.

Wait the two problems are apparently the price, and the reliability?

And you're asking a company famously burning money building a tool that is used for vibe-coding (aka unreliable software development) to build a replacement?

Idk man.

signal should just add better API / bot stuff and then we could all use that. there's no way OpenAI would be trustworthy for this; slack certainly isn't

My personal experience with using Slack as just a in-company chat app has been fine. I enjoy using Slack more than Teams or Discord.

All their integrations kinda suck though, and its not uncommon for integrations to randomly break with no discernible changes elsewhere.

They almost have, you could wrangle group projects + group chats together pretty easily and you'd be close-ish. The claude cowork experience backed by google drive with the openai group projects and group chat would, imo, be a really awesome way to work!

Slack's software quality has been in absolute freefall over the past couple of months.

Is it me or does the article mix characters with different fonts weights through the text?

I hate Slack. Total information overload. I’d prefer a tool that encourages people to think more before hitting send.

I for one would love this - if it’s done well - except that it would presumably be locked in to OpenAI agents

OT latent space podcast is great, most recently interview with jeff dean. Worth a listen

> Developers routinely complain about Slack’s API costs and permissions

What? What API costs is the op talking about?

They'll just do what Anthropic does: let it Ralph Wiggum a pile of broken shit, and then say "wewwwww, doing pwogwamming is vewwy hawd, UwU >_<" when it won't build and fails at basic use cases that would be easy to test automatically

If AI is soo productive why do they even sell it and don't hoard it for themselves to build a competing offer to everything ?