Thursday, November 6th 2025
Microsoft's AI Goal is Humanist Intelligence in Service of People and Humanity
Microsoft's head of AI Mustafa Suleyman made some bold promises in which direction the company's AI research will be headed. "At Microsoft AI, we're working towards Humanist Superintelligence (HSI): incredibly advanced AI capabilities that always work for, in service of, people and humanity more generally," notes the lengthy blog post, adding that "We think of it as systems that are problem-oriented and tend towards the domain specific. Not an unbounded and unlimited entity with high degrees of autonomy - but AI that is carefully calibrated, contextualized, within limits. We want to both explore and prioritize how the most advanced forms of AI can keep humanity in control while at the same time accelerating our path towards tackling our most pressing global challenges."
To achieve these goals, Microsoft has built a team called Microsoft AI (MAI) Superintelligence Team. Using massive amount of resources, combining human intelligence, hardware, software, and various other intelligence forms, Microsoft wants steerable AI systems. However, there is an issue. It appears that no one knows how will a superintelligent system be controlled. As Mustafa Suleyman claims, "No AI developer, no safety researcher, no policy expert, no person I've encountered has a reassuring answer to this question. How do we guarantee it's safe?"Instead of having a goal to surpass human at everything and take over, MAI Superintelligence team wants Humanist Superintelligence (SAI) to be the one that aids humans instead of replacing them. "Humanist superintelligence (HSI) offers an alternative vision anchored on both a non-negotiable human-centrism and a commitment to accelerating technological innovation… but in that order. The order is key. It means proactively avoiding harm and then accelerating. Instead of being designed to beat all humans at all tasks and dominate everything, HSI begins rooted in specific societal challenges that improve human well-being." Let us know your opinions about this, and whatever you believe Microsoft could build a human-centric AGI.
Sources:
MAI, via The Verge
To achieve these goals, Microsoft has built a team called Microsoft AI (MAI) Superintelligence Team. Using massive amount of resources, combining human intelligence, hardware, software, and various other intelligence forms, Microsoft wants steerable AI systems. However, there is an issue. It appears that no one knows how will a superintelligent system be controlled. As Mustafa Suleyman claims, "No AI developer, no safety researcher, no policy expert, no person I've encountered has a reassuring answer to this question. How do we guarantee it's safe?"Instead of having a goal to surpass human at everything and take over, MAI Superintelligence team wants Humanist Superintelligence (SAI) to be the one that aids humans instead of replacing them. "Humanist superintelligence (HSI) offers an alternative vision anchored on both a non-negotiable human-centrism and a commitment to accelerating technological innovation… but in that order. The order is key. It means proactively avoiding harm and then accelerating. Instead of being designed to beat all humans at all tasks and dominate everything, HSI begins rooted in specific societal challenges that improve human well-being." Let us know your opinions about this, and whatever you believe Microsoft could build a human-centric AGI.

53 Comments on Microsoft's AI Goal is Humanist Intelligence in Service of People and Humanity
I hope they opensource it, gpt-oss-20B style models tuned for specific domains would be quite nice.
The rest of the announcement is pure PR.
Microsoft hasn’t historically been the nicest of companies.
It’s been better behaved than say united fruit, union carbide and Monsanto but I wouldn’t say it’s ever been a nice and friendly or a producer of rock solid software.
- Either you control it meticulously and it quickly becomes a very expensive Clippy
- Or you let it go free and it will push the emotional to suicide, the average human to the mental asylum, and itself to pointlessness, while it eventually still reasons itself in a deep hole, nonetheless.
Sentience is far away. Its still a computer, and its still completely an affair of 'input = output'. AI doesn't do fuck all by itself. Its all human input, in the end.
Studies are showing that every AI you repeatedly ask similar questions, will shortcut its way to earlier answers. Minutely different, but similar, and not necessarily 'correct'. Basically the AI suffers from confirmation bias, just like humans do. Yet another strike against the AI: it reasons like humans do, because human efficiency dictates we aren't going to fix the same problem again, we just look back on how we did it before, AI is just as lazy as a human, at that, too. And it lacks the human drive, a motivator to keep exploring - so its effectively worse.
The bottom line, AI ain't better, its just a bigger dataset and it can distill something out of it. But in its practical use, its no more than an overpriced algorithm and once you've got the right one, you don't need the AI to do any learning anymore (but you could still call it 'AI'). This state is what Microsoft is describing here and I think its the only logical 'out' to somehow recover a snippet of the billions spent on it: distinct Agents that execute specific tasks - very expensive, but specialized Clippy's.
The main purpose of such a tool is that it can do things much faster and 24/7, which saves or will save money for large companies.
Just look at automated car factories as an example, some things are still done by humans, but many operations are automated, it will be the same here.
Another example, did you know that there are Microsoft 365 companions? You and I won't use it, but most people will just accept it. And they will improve this type of AI Apps, Linux will be an option, but for how long?
If only said drivel didn't make line go up...
Autonomous driving, much like any other tool just needs time to be refined! Yes it does something we already did and takes away a skill we used to need. But just wait! Soon! We already have cab rides in a controlled environment and sure its still a little problematic but just wait! Soon!
:toast:
And I wouldn't dare say that every technology company has so many stupid people in it that they would bet billions of $ on a "bubble."
And why do you think that "positive income" can only be "money"? ;)CBDC + AI management? :)
Most pressing global challenges are all the ongoing destruction of environment and various pollution.
Now how does wasting lots of electricity and all the hardware environmentally expensive to make from raw materials help to solve those?
Well, I guess if it's to make Skynet which then curbs this nature's top pest specie...
Certainly a lot of (poop) weight in M$ statement.
The sooner this AI cancer blows up the better.
I thing Goebbels is more credible then any effort from M$ PR spin-doctor carousel that spits those grandiose mission statements since 1995 without any shame
Oh man what a dude. You know, Steve's persistence might have leaked into Windows' update schemes and will never go away, like a piece of rogue code it just keeps trying to wedge its way into your life. Its the only possible explanation for all those applications that keep trying to set themselves as default...