Tired of ads? Enjoy an ad-free experience by signing up.
👩‍🍳 How we use AI at Tech in Asia, thoughtfully and responsibly.
🧔‍♂️ A friendly human may check it before it goes live. More news here

Microsoft launches superintelligence team to develop AI systems

Microsoft has launched the MAI Superintelligence Team, led by Mustafa Suleyman, to develop advanced AI systems after changes to its agreement with OpenAI that previously restricted its research.

The new team will focus on “humanist superintelligence,” a term Suleyman uses to describe AI that serves people and society.

Microsoft’s previous deal with OpenAI had limited its ability to train large models and conduct independent AGI research.

The company now plans to pursue its own superintelligence work while continuing collaboration with OpenAI through at least 2030.

Karén Simonyan, who joined Microsoft with Suleyman from Inflection, will serve as chief scientist for the new team, which also includes researchers from firms like Google DeepMind, Meta, OpenAI, and Anthropic.

🔗 Source: Fortune

🧠 Food for thought

Implications, context, and why it matters.

Amended OpenAI deal lets Microsoft run its own AGI research and train large models

  • The earlier OpenAI deal had ‘limited its ability to train large models and conduct independent AGI research’ for Microsoft. The new pact lets Microsoft ‘pursue its own superintelligence work’ and restores the right to train large language models on its own.
  • UBS says the pact keeps Microsoft’s ‘exclusive IP rights and Azure API exclusivity until AGI is achieved’ 1. It also extends IP and hosting rights for OpenAI models to 2032 1. UBS pegs Microsoft’s commitment at about $250 billion over seven to eight years 1.
  • Microsoft formed the MAI Superintelligence Team led by Mustafa Suleyman with researchers from Google DeepMind plus Meta, OpenAI and Anthropic. The effort builds parallel AI work, while Microsoft keeps its roughly 27% OpenAI stake 2.

Dual-track AGI plan boosts compute for cloud and chip firms

  • MAI efforts run alongside $30–$35 billion in yearly OpenAI commitments, per UBS 1, which lifts compute needs and opens deals for datacenter builders, power suppliers plus cooling vendors.
  • Wisconsin construction features an AI site with hundreds of thousands of Nvidia Blackwell GB200 GPUs 3, and a second site planned for 2027 or later would push power needs above 900 megawatts 3.
  • Microsoft has built custom silicon for AI 4, so Nvidia rivals can pitch training parts for MAI workloads.

Recent Microsoft developments

Stay ahead in Asia’s tech landscape

You've reached your 2 free content limit for the month. Sign up for free to read the full story.

🏄 For casual readers / 👶 Free

Basic

US$0

Free forever

Get instant access to this article and more every month

0 premium content

Unlimited news briefs

5

5 articles

Ad-free reading experience

Just US$0 per day

⌛Sign up in 20s. No payment details needed.

📖 For learners / 👍 Starter

Lite

US$4.92/month

Billed annually at US$59/year

Get instant access to this article and more every month

4

4 premium content

Unlimited news briefs & articles

Ad-free reading experience

Just US$0.17 per day

Cancel anytime

Our subscriber community includes professionals from these companies:

Stay updated on the go with our mobile app.

Get latest insights with smoother, more personalized experience through TIA mobile app.