👋 Tomorrow’s Tech, Delivered Today

Hi! Welcome to the 52nd edition of the TomorrowToday newsletter.

We’re here to decode the AI chaos so you don't have to. Think of us as your friendly neighbourhood tech translators - we cut through the chaos, translate the jargon, and spotlight new AI tools that matter for founders, builders, and curious minds.

Buckle up, because the future's moving fast and we're here to make sure you don't get left behind! ⚡

If you enjoyed today’s newsletter, please forward it to a friend & subscribe by following this link.

~5 mins read

🗞️ News Flash

🤖 Microsoft Scout: The always-on agent that runs your work life while you're in meetings

/Productivity /Agents /Microsoft

Microsoft just announced Scout at their Build 2026 conference, and it's the most significant shift in how they think about AI at work since Copilot launched. Scout isn't a chatbot you open and ask questions, it's an always-on personal agent that lives inside your Microsoft 365 ecosystem and quietly gets stuff done on your behalf, 24/7.

Here's what makes it different. Scout connects to Teams, Outlook, OneDrive, SharePoint, and your calendar, and uses something Microsoft calls Work IQ - an intelligence layer that maps out how you work: who you collaborate with, what projects are active, and where things are stalling. Instead of waiting to be asked, Scout proactively resolves scheduling conflicts, preps you for meetings, and joins Teams group chats as an actual participant (not a sidebar tool). Think of it less like a search bar and more like a very switched-on colleague who's always one step ahead.

Microsoft CEO Satya Nadella introduced it as the first of a new product category called "Autopilots", a new class of agents that act with their own identity and work on your behalf. It's currently in experimental preview for enterprise Frontier customers, with a broader rollout planned for late June and general availability expected around October 2026. Pricing is still to be confirmed, but Microsoft has indicated Scout will be an add-on for M365 E3 subscribers and included with E5.

We really hope that this is going to be a good feature because in the past Microsoft has had a tendency to overpromise and underdeliver.

Real-life use case: Ask Scout to prep a briefing before your next big meeting by scanning your emails, calendar, and Notion docs - and come back to a ready-made summary.

Google Spark: The same idea, but for your entire Gmail life

/Productivity /Agents /Google

Not to be outdone, Google launched Gemini Spark this week, and if Scout is Microsoft's answer to the agent race, Spark is Google's. The parallel is almost too on the nose: both launched within days of each other, both promise to be your always-on AI agent, and both are betting that whoever owns your inbox owns your workflow.

Spark runs inside your Google Workspace - Gmail, Drive, Calendar, Docs - and operates in the background, even when your laptop is off. You give it a task and it runs. Ask it to scan your inbox every Monday at 9am, summarise the week's most important emails, build a prioritised to-do list, and block time on your calendar for deep work, and it does exactly that, automatically. You can even teach it a custom skill called "ghostwriter" by pointing it at your last 50 emails and asking it to learn your writing style. It then uses that skill every time you ask it to draft something.

The secret weapon here is obvious: Google has your entire digital life. Your Gmail, your Drive, your search history, your calendar. In the agent race, data is the moat, and Google's moat is very, very wide. Spark is currently marked as "coming soon" for Gemini Ultra subscribers in the US, but it signals clearly where things are heading.

Real-life use case: Set up a recurring Monday morning digest - Spark scans your inbox, summarises the week, and schedules deep work blocks so you can start the week ready, not reactive.

💻 Google Gemma 4 12B: A powerful AI model that runs entirely on your laptop

/OpenSource /LocalAI /Google

Most of us access AI through subscriptions, Claude, ChatGPT, Gemini, and that model works brilliantly. But Google just released something interesting for a different crowd: Gemma 4 12B, a free, open-source AI model that runs entirely on your laptop, no internet required.

For most readers, this might sound like a developer story. But the real-world implications are significant. If you're on a flight, in a load-shedding blackout, or working with sensitive data you'd rather not send to the cloud, a local model like Gemma becomes genuinely useful. It runs on any machine with 16GB of RAM (including most modern MacBooks with Apple Silicon), handles text, images, audio, and video, and, crucially, performs close to models twice its size in standard benchmarks. It's free to download and use forever, no subscription required.

Google released it under an Apache 2.0 licence, meaning anyone can download, modify, and build on top of it. Gemma 4 12B is Google's clearest signal yet that the future of AI isn't only in the cloud, it's increasingly on the device in your bag. If you're curious, it's available on Hugging Face and can be run through tools like LM Studio with minimal setup.

Real-life use case: Run a powerful AI assistant offline on your laptop - perfect for flights, load-shedding, or keeping sensitive business data off the cloud entirely.

💡 Curiosity Corner

In this section, we aim to spotlight an incredible AI tool or use case and guide you on how you can try it.

This week’s challenge: Discover your ideal career path with Google's Career Dreamer

Google quietly released a whole set of little AI-powered mini apps - and one of them is genuinely worth 10 minutes of your time. It's called Career Dreamer, and it's an interactive AI experience that helps you figure out what you're naturally suited for, what careers align with your strengths, and what skills might unlock new opportunities for you.

It doesn't feel like filling in a form. It feels like a conversation. Whether you're a student figuring out a direction, someone thinking about a career pivot, or just curious about where your strengths might take you - it's a surprisingly thoughtful tool.

Here's how to try it:

  1. Answer a few questions about your background, interests, and strengths - it's more of a conversation than a quiz

  2. Explore the careers it suggests and how they map to your current skills

  3. Click into any career to see what skills you'd need, what learning paths exist, and how far you already are

No sign-up required. It's free, takes about 10 minutes, and might just prompt a very interesting conversation with yourself.

🏢 AI in Enterprise

In this section, we're spotlighting real businesses using AI to solve actual problems.

The AI That's Learning to Build Itself

This week, Anthropic, the company behind Claude, published something that genuinely stopped us in our tracks. It's not a product announcement. It's more of a quiet alarm bell. The piece is titled "When AI Builds Itself," and it documents something that sounds like science fiction but is apparently already happening inside one of the world's leading AI labs.

Here's the core of it: today, Claude writes more than 80% of the code that gets merged into Anthropic's own systems. Not assisting engineers - writing it. That number was in the low single digits before early 2025. In the same period, Anthropic's engineers now ship roughly 8 times as much code per quarter as they did in 2024. The AI isn't replacing the engineers; it's making each one dramatically more productive. But Anthropic is honest that the human role is shrinking with each iteration.

The term they're using is recursive self-improvement. The idea that AI systems could eventually design and train their own successors, without needing humans at every step. We're not there yet. But Anthropic is saying the trend lines suggest it could arrive sooner than most organisations, governments, regulators and companies are prepared for.

What does this mean for you? In the short term, not much changes. AI is still a powerful tool that does what you tell it. But the rate of improvement is genuinely accelerating. Tasks that took AI models an hour to complete in early 2025 now take about 12 hours , meaning the ceiling for what AI can handle autonomously keeps rising. Anthropic's call to action is clear: the institutions that shape how this technology develops need to start paying attention now, while humans are still firmly in the loop.

Anthropic is also calling for a slowdown in the development of frontier AI. Whether this is, like they say, for the good of humanity and to make sure that alignment research catches up with the frontier models, we don't know. Or maybe it is a ploy to make sure that the entire industry and Anthropic competition slows down to keep them in the lead of a competitive advantage.

What the true motivation is no one will know. All that we know is that AI is improving faster than ever.

📜 AI Dictionary

AI is full of jargon, and we’re here to decode it. Each week, we’ll give you a plain-English definition of a buzzy term you’ve probably seen (but never fully understood).

Agentic Workflow - noun

When AI doesn't just answer your questions - it actually goes and does things. Instead of you running each step yourself (open email, draft reply, save to Drive, update the calendar), an agentic workflow chains those actions together automatically. You give the AI a goal; it figures out the steps and executes them. Scout and Spark, covered in this week's News Flash, are both agentic workflows living inside your work apps. Think of it as the difference between asking someone for directions and handing them the keys.

We’d like to ask a favour 🤝
If this email lands up in your Promotional or Spam folder, please move it to your Primary inbox. We’re working hard to bring you the best content weekly, and your support is truly appreciated. Thanks!

Thanks for reading TomorrowToday! We’d love to hear from you:

➡️ What would you like us to cover next?
➡️ Have a tool or topic we should feature?

We’re building this with (and for) you. 🚀
See you next Tuesday 👋

Keep Reading