Microsoft’s new direction: value-driven superintelligence

Microsoft has announced the formation of its new “MAI Superintelligence” team, led by Mustafa Suleyman, which will focus on developing advanced AI systems explicitly designed to serve human priorities, rather than merely achieving technological milestones. 

Suleyman described the shift as: “We are not building an ill-defined and ethereal superintelligence; we are building a practical technology explicitly designed only to serve humanity.”

The emphasis is on AI tackling concrete global challenges—such as healthcare, clean energy, and education—while maintaining robust human control and ethical alignment.

Why it matters

• By framing its superintelligence efforts as “humanist”, Microsoft signals a pivot away from competitive race-to-AGI narratives and toward value-driven applications.

• This signals to enterprise customers and regulators that AI advancement can align with societal goals—not just raw capability.

• For prompt-engineered creators and platforms like ours, it suggests a greater demand for high-quality, thoughtfully framed AI workflows that reflect real-world value and ethical context.

EU weighs delay on landmark regulation

At the same time, the European Commission is reportedly considering a delay in the rollout of key provisions of the EU AI Act. 

Under pressure from major tech companies and the U.S. government, Brussels may grant a one-year “grace period” for firms developing generative AI systems and defer enforcement of transparency or transparency-related fines until 2027.

That said, the Commission emphasises that no decision has yet been made and remains “fully behind the Act and its objectives”.

What it means

• Regulators may be signalling flexibility in the face of innovation pressure—which could accelerate AI deployment in Europe.

• Vendors and prompt practitioners should prepare for both increased demand and evolving compliance expectations: high-risk use-cases will face scrutiny, but interim leeway may open opportunities.

• For our prompt-library business, this means we could highlight “compliance-aware” workflows and prompts optimised for use-cases that may fall under regulation. This positions us ahead of the curve while regulation crystallises.

Prompt Tip of the Day

Prompt: “Act as a ‘human-centred superintelligence advisor’. First identify an overlooked global challenge (e.g., clean energy or equitable education). Then design a 3-step AI workflow for addressing it, emphasising human values, measurable impact, and minimal regulatory risk.”

Use this in your next brainstorming session—whether with ChatGPT, Claude or Gemini—to shift focus from “what AI can do” to “what AI should do”.