Nine months.
That's how long it took Manus AI to go from being called an "unusable shell product" to closing a $2 billion acquisition with Meta.
This is the fastest failure to acquisition story in AI history. And it tells us everything about where this industry is heading.
The Fall
March 2025. Manus drops a demo that breaks the internet.
The pitch? The first "general AI agent," an AI that actually does things, not just talks about them. You give it a task. Watch it plan. Watch it execute. Get real results.
Invite codes sold for $14,000. People called it the next DeepSeek moment.
Then reality hit.
Chinese social media turned hostile. The accusations were brutal:
"Pure shell product," just a wrapper on Claude
Users burning 800 credits uploading a single file
Reddit threads calling it "absolutely unusable"
Most startups die here. Once the "scam" narrative takes hold, you can't outmarket your way out.
But Manus didn't die.
The Rise
Instead of chasing Twitter virality, they started chasing invoices.
June 2025: Playbook templates cut credit consumption by 30%.
July 2025: Chat Mode made simple queries cost 5 credits instead of 100.
Then enterprise deals started landing. Mercury Law reported 40% time savings on contract review. Shanghai Zenith Advertising adopted it for campaign automation.
Real companies. Real invoices. Real validation.
By December 2025: $100 million ARR. Two million users.
From "shell product" to nine figure revenue in six months.
That's not a pivot. That's a resurrection.
Why Meta Paid $2B in 10 Days
Normal M&A takes months. Meta closed in 10 days.
That's not strategy. That's desperation.
Here's what Meta was facing:
Google rolling out Gemini Agents across Workspace
Microsoft embedding Copilot everywhere
OpenAI shipping function calling agents rapidly
Anthropic's Claude becoming the go to for agentic workflows
Everyone was moving. Meta was stuck.
And here's what nobody's talking about: Meta is quietly shifting away from open source. They're building a proprietary model codenamed "Avocado."
The "Llama is free forever" era? That's changing.
Meta didn't buy revenue. They bought an agentic architecture they were years behind on.
The Takeaway
Chatbots chat. Agents act.
That's the paradigm shift. Gartner says 40% of enterprise apps will have AI agents by end of 2026, up from 5% the year prior.
The question isn't "how do I make my chatbot smarter?"
It's "how do I make my AI actually do things?"
Big Tech is paying billions to answer that question. Are you?
Reply to this email and tell me: are you building with AI agents? I read every response.
Love this newsletter? Share it with a friend and unlock exclusive rewards.
Until next time,
Steve Oak