Remember when Slack was just a place to send GIFs and pretend you were working? Those days are officially over. Salesforce just announced they’re cramming 30 new AI features into Slack, and I have one question: did anyone actually ask for this?
Look, I’ve spent the last three years reviewing AI tools, and I’ve developed a finely-tuned BS detector. When a company announces 30 features at once, that’s not innovation—that’s a feature dump. It’s the corporate equivalent of throwing spaghetti at the wall and seeing what sticks.
What We Actually Know
Salesforce rolled out these changes in 2026, building on their January update that gave Slackbot “agentic capabilities.” For those keeping score at home, that’s corporate speak for “the bot can now do stuff on its own.” The company held a small gathering in San Francisco to announce this AI-heavy makeover, which tells you everything you need to know about their confidence level. Big innovations get big stages. Incremental updates get intimate gatherings.
The pitch is simple: these features will integrate AI deeply into work workflows. They’re positioning the new Slackbot as a personal AI agent that’s grounded in your company’s data, workflows, and Slack conversations. It’s supposed to bring AI “right into the flow of work.”
The Real Question Nobody’s Asking
Here’s what bothers me: Slack already had a problem, and it wasn’t a lack of AI. The problem was that Slack had become a chaotic mess of channels, threads, and notifications that made actual work harder, not easier. Adding 30 AI features doesn’t solve that problem—it just adds 30 more things to learn, configure, and inevitably ignore.
I’ve watched this pattern play out dozens of times. A company buys another company (Salesforce bought Slack for $27.7 billion, in case you forgot), then feels compelled to justify that purchase by bolting on features from their existing product line. It’s not about making Slack better. It’s about making Salesforce’s AI investment look smart.
The Agentic Capabilities Trap
Let’s talk about these “agentic capabilities” for a second. The idea of an AI agent that can act autonomously in your workspace sounds great until you think about it for more than five seconds. Do you really want an AI making decisions about your work communications without your direct input? Do you trust it to understand context, tone, and office politics?
I’ve tested enough AI agents to know that “grounded in your company’s data” is doing a lot of heavy lifting in that sentence. These systems are only as good as the data they’re trained on, and most company Slack instances are a dumpster fire of inside jokes, half-finished thoughts, and messages that made sense at 2 AM but look insane in the morning light.
What This Really Means
Salesforce is remaking its entire business around AI. That’s not news—every tech company is doing the same thing. But there’s a difference between thoughtfully integrating AI where it adds value and just spraying AI features everywhere because that’s what investors want to hear about.
The features are rolling out “in the coming months,” which is another red flag. If these 30 features were truly ready and truly valuable, they’d launch them all at once with specific dates. Vague timelines usually mean the features aren’t quite baked yet, or they’re hedging their bets in case some of them turn out to be duds.
My Take
I want to be wrong about this. I genuinely do. Slack is a tool millions of people use every day, and if Salesforce can make it meaningfully better with AI, that’s a win for everyone. But 30 features at once isn’t a strategy—it’s a spray-and-pray approach that suggests they’re not entirely sure what will work.
The best AI tools I’ve reviewed are focused. They do one thing exceptionally well, or they add AI in ways that feel natural and unobtrusive. Announcing 30 features suggests the opposite approach: throw everything at the problem and hope something resonates.
Will some of these features be useful? Probably. Will all 30 be worth your time? Absolutely not. And that’s the problem with this kind of announcement. It creates noise instead of clarity, and it puts the burden on users to figure out which features actually matter.
So yeah, Slack is getting an AI makeover. Whether that makeover is a glow-up or just expensive cosmetic surgery remains to be seen. But based on what we know so far, I’m not holding my breath.
đź•’ Published: