OpenAI’s Leaked Internal Document Outlines “Super Assistant” Mission

Good morning. It’s Monday, June 2nd.

On today in tech history: 2014: Apple introduced Swift, a statically-typed, LLVM-compiled language for iOS/macOS apps, featuring type inference, null-safe optionals, and ARC. As much as 2.6x faster than Objective-C, its secure, readable syntax and 2015 open-source release (Apache 2.0) with SwiftUI support revolutionized scalable, cross-platform app development.

OpenAI’s internal doc leak
Google releases an app that helps you to run AI models locally
Perplexity pushes into enterprise tools with Labs release
5 Latest AI Tools
Latest AI Research Papers

You read. We listen. Tell us what you’re thinking that by replying to this email.

Meet Suna — Your First AI Worker

Forget assistants. Hire your first AI worker.

Suna doesn’t just answer questions. It takes motion in your behalf.
Need 20 investor contacts? A visit planned across continents? A competitive landscape in PDF format by morning? Suna does that.

Not “summarizes.” Not “suggests.”
Does.

✅ Scrape the net for candidate resumes
✅ Generate personalized sales outreach
✅ Analyze markets, web sites, web optimization
✅ Cross-reference papers, construct reports, send spreadsheets
✅ Plan entire company trips across borders

Suna is essentially the most capable generalist AI agent we’ve seen — and it’s 100% open source.

Built by Cavtixtrusted by developers, and now available to anyone. Get 15% off if you use the code AIBREAK at checkout.

You don’t prompt Suna.
You delegate to it.

^{Thanks for supporting out sponsors!}

Today’s trending AI news stories

OpenAI wants ChatGPT to be a ‘super assistant’ for each a part of your life

OpenAI is about to release the o3-pro model soon, having already alerted select customers to its imminent release. This advancement forms a key part of a bigger strategy detailed in a leaked internal document. It sketches out a future where ChatGPT, powered by the upcoming o3 model, becomes a full-spectrum operator: managing calendars, booking travel, navigating software, and even contacting professionals in your behalf.

It refers back to the assistant’s capabilities as “T-shaped”: broad enough to administer each day tasks, yet deep enough to handle specialized work. A generative UI is paired with tools like “Computer Use,” giving the assistant hands-on control over devices and applications. OpenAI can also be constructing its own search index, letting ChatGPT sidestep traditional engines and act more like a one-stop interface than a passive responder. Despite redactions, the memo further indicates that hardware integration stays central to this direction.

For now, monetization takes a back seat, with OpenAI prioritizing usage growth to trial latest revenue models within the second half of 2025, including potential business models at no cost users. Read more.

Google quietly released an app that helps you to download and run AI models locally

Google AI Edge Gallery enables Android users to download and run open-source AI models locally, using on-device processing to perform tasks comparable to image generation, code writing, and query answering without requiring a web connection. Compatible with models like Gemma 3n and featuring a configurable “Prompt Lab” for single-turn prompts, the app’s performance is dependent upon device hardware and model size, with an iOS version forthcoming.

The welcome screens of the Google AI Edge Gallery app for Android. Image: Google

As Veo 3 rapidly generated thousands and thousands of AI videos across 71 countries, driving TPU infrastructure to capability, Google emphasized responsible AI use by watermarking all Veo 3 videos, aside from Ultra-tier users of the Flow tool, which supports specialized workflows within the Gemini app on Android and iOS. Planned updates will introduce image-to-video conversion, improved audio fidelity, faster rendering, and integration with Google Workspace, underscoring Google’s commitment to combining local AI capability with scalable cloud-based production tools. Read more.

Last Friday, we shipped Veo 3 to 71 latest countries, Pro members, and Ultra members got more credits. All week we have been scrambling to maintain every thing up and running – way, way, far more demand than we expected!

Today, 2 more updates:
+ The UK now has Veo 3 access 🇬🇧
+ Pro and

— Josh Woodward (@joshwoodward)
6:57 PM • May 30, 2025

Perplexity pushes into enterprise tools with Labs release

Perplexity has launched Perplexity Labsa productivity tool for Pro subscribers ($20/month) that generates reports, spreadsheets, dashboards, and interactive web apps using AI. Available on web, iOS, and Android, with desktop apps to follow, Labs handles prolonged tasks, using features like code execution, chart creation, and file generation. Results, including images and documents, are organized in a dedicated workspace for download or review.

The launch signals Perplexity’s pivot beyond search, arriving the identical day as Manus debuted its AI slide generator. Labs supports data structuring, formula application, and custom workflows, all powered by an AI system that typically takes 10 minutes or more to finish a task. The corporate can also be testing an internet browser (Comet) and recently acquired skilled social platform

This broader product push aligns with Perplexity’s corporate focus, following the launch of an enterprise plan and reported efforts to boost as much as $1B at an $18B valuation. Read more.