Launching Copilot Tasks, a new way to get stuff done! Tasks is the best of models with access to the best of Microsoft (Edge, Browser, Infra). It's your consumer grade clawdbot but on cloud.
The models are here, and Microsoft has the tools to make the most powerful personal productivy agent. You can build docs, slides and sheets OR ask it to do real world chores you hate to do yourself. This is what a personal assistant should be.
Go to https://t.co/2uiiq0gLTM to get on the waitlist!
Super excited to announce seven new world-class MAI models today. They represent what we consider a new era in AI designed to keep you in control and on the frontier.
First is our text foundation model, MAI-Thinking-1, exceptionally strong on reasoning and SWE tasks.
- It’s a 35B active parameter MoE with a 256K context window. Independent human raters on Surge prefer it for overall quality in blind side-by-sides versus Sonnet 4.6, and it’s achieved 97% on AIME 2025, the key measure of its general-purpose reasoning abilities.
- It's at 53% on SWE Bench Pro, placing it right alongside Opus 4.6 on one of the toughest coding benchmarks.
- And since we co-designed our models with our own silicon, MAI-Thinking-1 is optimized on our MAIA 200 chip. Benchmarking head-to-head against the GB200, we see 30% better performance per dollar as well as a 1.4x performance-per-watt gain when running our MAI models on the MAIA 200 end-to-end.
Next is MAI-Image-2.5 and its Flash variant. Two super strong models now at #2 on the leaderboards, surpassing the score of Nano Banana 2 on image editing.
Last for now is MAI-Code-1-Flash, our new inference efficient coding model, especially tuned for VS Code and GitHub Copilot CLI.
- Code-1-Flash achieves 51% on SWE Bench Pro, despite having just 5B parameters, putting it closer to Haiku in size but cheaper in cost.
All of this is the foundation for Microsoft Frontier Tuning. It lets you customize our models to create custom, company-specific agents that only you control. You can make our model, your model. Your data. Your agents. Your moat.
Early adopters are already seeing a difference. When we tuned our models for McKinsey’s tasks, MAI delivered the highest win rate, outperforming GPT-5.5 on quality, while being 10x lower on cost.
Also really excited to be collaborating with the amazing team at Mayo Clinic to jointly train a new frontier AI model for healthcare.
Our announcements today mark another milestone on the road to humanist superintelligence. You can learn more and about our other new models in our latest blog: https://t.co/v65eop5Ixq
Don't make a sliver of a product .. make it feature-complete without compromising on speed to win against incumbents
Go after the hard problem today, you'll be surprised by how we are totally miscalibrated on what's hard
Founders must stop trying to building 2010-era businesses with 2026-era technology.
Don't try to rebuild Foursquare or Yelp.
Don't try to recreate Basecamp by 37 Signals with $10/mo SaaS pricing.
Don't underprice! If it works it's worth a lot more.
Don't be tempted to become "Tech enabled PE" with revenue tricks.
The rules of tech changed with AI. Play the new game.
Skip the busywork. Copilot Tasks helps you stay on track by taking care of the tasks that slow you down.
Try Tasks & join the waitlist today: https://t.co/ipVeDXHU0r
AI shouldn’t be confined to pilots and side experiments.
That’s why Accenture is scaling Microsoft 365 Copilot to nearly its entire global workforce – 743,000 people across Accenture and Avanade – and embedding AI directly into everyday work.
Based on 2025 company data from the first 200K employees using Copilot, they are completing routine tasks up to 15x faster, seeing a 53% improvement in productivity and efficiency, and driving strong adoption across the organization.
Learn how Accenture is turning AI into a real work assistant at enterprise scale: https://t.co/tLNQxf59E5
Super excited GPT-5.5 is rolling out to GitHub Copilot, M365 Copilot, Copilot Studio, and Foundry today.
With deeper reasoning, stronger multistep execution, and better performance across long, complex tasks, GPT-5.5 helps you go from idea to execution faster with fewer iterations to get to the right outcome.
It’s all about helping you choose the right model, or models, for the right task across your workflow.
I've been using @Copilot Tasks for travel planning and it has been a GAME CHANGER. It connects to your Email, One Drive, Google Drive etc. and with a simple prompt - Tasks creates a travel plan with recommendations, actions, suggestions! This is awesome @Microsoft@satyanadella
Tax season can be... a lot. Copilot Tasks helps by digging through your email for the right forms, finds what you're missing, and puts it all in one place so you can get it done with a little less chaos.
Try Copilot Tasks today: https://t.co/zoW2HcfGNx
Using Copilot Tasks to answer all my tax advisors questions has been phenomenal. Here's a fun email I got:
"For FBAR we need all Indian account details for you and your wife with their maximum balance in 2025"
My prompt to Copilot Tasks.
"My CA sent me an email asking for stuff. Can you compose a reply with the details from my drive and email accounts"
Found documents from drive and emails my wife sent me, unlocked the PDFs(with my consent and info), reasoned to get 2025 max values, made and email, and I hit send.
🤯🥹 Try now at https://t.co/Runfhn2xNR