Dorsey makes the AI case against managers

⋅

Read Online | Sign Up | Advertise

Good morning, {{ first_name | AI enthusiasts }}. Every big company runs the same way: stack managers on top of managers until information flows. Jack Dorsey thinks AI just made that whole structure obsolete.

The Twitter founder and Block CEO just laid out a plan to replace his company’s entire management layer with AI, coming weeks after cutting nearly half of its workforce and betting a smaller, flatter team can move faster than any org chart ever could.

Reminder: Our next live workshop is today at 2 PM EST! Join and learn about the different AI tools for presentations out there, and get a pre-build workflow to leverage the tech to do the heavy lifting on your own slides. RSVP here.

In today’s AI rundown:

  • Block ditches managers for AI

  • SpaceX targets record $1.75T IPO debut

  • Build a productivity tool with Replit

  • OpenAI taps freelancers to teach ChatGPT their jobs

  • 4 new AI tools, community workflows, and more

LATEST DEVELOPMENTS

BLOCK

🏗️ Block ditches managers for AI

Image source: Lovart / The Rundown

The Rundown: Twitter founder and Block CEO Jack Dorsey just co-authored a post arguing AI can replace middle management, framing Block’s recent 40% workforce cut as the opening move in a massive workplace restructure for the AI era.

The details:

  • Block cut over 4K employees in February, over 40% of its staff — with Dorsey calling it a bet on AI, not a response to weakness.

  • Dorsey said managers exist to route information up and down a chain, and AI can now do that via a live “world model” of the business.

  • He said everyone at Block now falls into one of three roles: builders, problem-owners over specific outcomes, and player-coaches who develop talent.

  • Block is remote-first, and Dorsey says every decision, design, and plan already exists as a digital record, giving AI the raw material to replace managers.

Why it matters: Dorsey’s thesis is an interesting one, especially as lean, AI-first teams go head-to-head with bloated legacy firms that have layers of approval. Block’s bet is that remote work already generated the data, and AI just needed to catch up to use it — but not everyone is going to trust the tech to completely cut out the managerial layer.

TOGETHER WITH HUBSPOT

💸 Turn AI into your income engine

The Rundown: HubSpot’s new “200+ AI-Powered Income Ideas” free guide offers actionable strategies to turn artificial intelligence into your own personal revenue generator — unlocking a gateway to financial innovation in the digital age.

With this guide, you can:

  • Explore hundreds of revenue-generating ideas across industries with real-world applications

  • Follow simple, step-by-step instructions that make AI accessible to everyone

  • Adopt cutting-edge strategies to keep you ahead in today’s fast-paced market

Access your free guide here.

SPACEX

🚀 SpaceX targets record $1.75T IPO debut

Image source: Lovart / The Rundown

The Rundown: SpaceX just filed for what would be the largest IPO in history, targeting a valuation north of $1.75T and a raise of up to $75B — which would make Elon Musk’s rocket-AI-social media mega-company one of the most valuable on Earth.

The details:

  • The SEC filing sets up a June debut that would beat OpenAI and Anthropic to public markets, making Musk’s company the first U.S. AI-era mega-listing.

  • SpaceX is targeting a $1.75T+ valuation, and its $50B–$75B raise would more than double the largest IPO ever (Saudi Aramco’s $29B offering in 2019).

  • Musk absorbed xAI into SpaceX before filing, though the AI side reportedly pulls in under $1B in revenue against the rocket business’s roughly $20B.

  • About 30% of shares would be open to everyday investors, while a special two-tier voting structure lets Musk keep full control after going public.

Why it matters: After all of the talk surrounding AI mega-IPOs centering on OpenAI and Anthropic, it’s xAI (via SpaceX) that will be the first U.S. lab to hit the public markets. Despite now losing every one of his 11 co-founders, Musk’s vision and tie-in of rockets, AI, robotics, and data make for a combo few other rivals can match at scale.

AI TRAINING

🧑‍💻 Build a productivity tool with Replit

The Rundown: In this guide, you will learn how to build a lightweight work tracker that counts what you actually did each day and turns it into a clean weekly report.

Step-by-step:

  1. Open Claude or ChatGPT and ask it to interview you about your role. You want to find the 5 to 8 task types you repeat most often. Good outputs are things like sales calls, deliverables sent, client replies, internal meetings, or content published.

  2. Take that task list into Replit and tell it: Build me a simple tracker app with daily number inputs, an optional note field, a calendar heatmap, and a report generator for any date range.

  3. Log a few fake days first so you can see the app working right away.

  4. Click Generate Report and have the app output totals, daily averages, and a short summary for the week or month. This is the part that makes the tracker useful. The data is already formatted for a manager update, client recap, or team check-in.

  5. Deploy the Replit and save the url as a bookmark. Start using this every day at the end of your workday.

Pro tip: If you want to make it more powerful, add a client or project tag to each entry. That gives you filtered reports by account instead of one big pile of activity.

PRESENTED BY UNWRAP

⚡ Powerful insights for powerful brands

The Rundown: Unwrap aggregates all your customer feedback (surveys, reviews, support tickets, social, sales calls, etc.) into a single AI-powered view, helping product, support, and CX teams at Southwest Airlines, Stripe, Lululemon, and DoorDash turn every customer signal into actionable insights.

With Unwrap, you get:

  • All customer feedback auto-categorized into a single view

  • Natural language queries to explore feedback instantly

  • Real-time alerts, custom reporting, and clear sentiment tracking

Connect with Unwrap to get a free trial of the tools, exclusive to The Rundown AI readers.

OPENAI

🎭 OpenAI taps freelancers to teach ChatGPT their jobs

Image source: Lovart / The Rundown

The Rundown: A new report from Business Insider just revealed “Project Stagecraft,” an internal OpenAI effort paying as many as 4K freelancers at least $50/hr to build occupation-specific training data across a variety of jobs.

The details:

  • The project runs through Handshake AI, with freelancers from jobs including commercial aviation, pharmacists, plant scientists, and HR specialists.

  • The project focuses on “knowledge work, not manual labor,” aiming to map economically relevant tasks and gauge what ChatGPT can already handle.

  • Contractors create personas and simulate workflows, providing “context, goals, references, and deliverables” to help train models with human expertise.

  • One contractor who participated told BI, “We all were aware that we were basically training AI to replace us.”

Why it matters: AI training has gone from generalist data labeling to a more targeted cataloging of what professionals actually do, field by field, task by task. With OAI also drafting policy papers on economic disruption and “rethinking the social contract,” the AGI timelines may be going much faster than even they anticipated.

QUICK HITS

🛠️ Trending AI Tools

  • ⚙️ Strands Agents – Innovation AI for everyone, from production backends and physical robots to code that writes itself*

  • 🧠 Trinity-Large-Thinking – Arcee AI’s new SOTA open-weight reasoning model for long-horizon agentic tasks

  • 🤏 LFM2.5-350M – Liquid AI’s small model built for tool use and on-device agents

  • 🎨 Wan2.7-Image – Unified image model for generation, editing, text rendering

*Sponsored Listing

📰 Everything else in AI today

Contra Labs emerged from stealth as a new evaluation platform for AI creative tools, with leaderboards, datasets, and benchmarks focused on human creative taste.

Z AI rolled out GLM-5V-Turbo, a new ‘vision coding’ model that reads screenshots, design drafts, and interfaces to generate runnable code directly from what it sees.

Liquid AI released LFM2.5-350M, a small open model that outperforms models twice its size on tool use and is able to run efficiently across consumer devices.

Arcee AI introduced Trinity Large-Thinking, a new open-weight reasoning model rivaling Opus 4.6 on agent benchmarks at roughly 1/20th the cost.

Alibaba launched Wan2.7-Image, a new image model that generates, edits, and renders text across 12 languages with up to 12 consistent images per prompt.

COMMUNITY

🤝 Community AI workflows

Every newsletter, we showcase how a reader is using AI to work smarter, save time, or make life easier.

Today’s workflow comes from reader Thomas L. in France:

“We have a Discord server with friends where we discuss everything from video games to politics. Conversations can sometimes get heated, especially when opinions clash.

To help with that, I built a Taoist-inspired bot powered by Claude. When mentioned, it reads recent messages, uses context about participants and our shared vocabulary, and sends a prompt to Haiku via API.

It then returns a response meant to bring perspective, clarity, and occasionally a bit of humor!”

How do you use AI? Tell us here.

🎓 Highlights: News, Guides & Events

See you soon,

Rowan, Joey, Zach, Shubham, and Jennifer — the humans behind The Rundown

Liked Liked