Microsoft wants to put a dollar value on AI’s work – but can you trust the math?

28 minutes ago 3

ZDNET’s cardinal takeaways

  • Microsoft’s caller instrumentality estimates the concern worth of AI agents.
  • It compares AI costs against the worth of quality labor.
  • Estimated economical worth doesn’t needfully mean existent savings.

We each cognize AI tin get expensive. But does that disbursal effect successful existent concern value? That’s the trillion-dollar question Microsoft aims to reply with today’s instauration of Agent 365 Business Analytics.

Also: Microsoft says its caller AI coding exemplary rivals Claude Opus 5.5 astatine 5% of the cost

How overmuch worth successful existent dollars are your agents producing?

If it works, this is immoderate precise chill stuff. Microsoft is trying to representation labor-market economics onto AI usage information to make worth estimates.

Also: Work IQ is Microsoft’s large stake connected agent-first endeavor IT, and I person questions

To bash so, it uses existent labor-costing information, including information from the US Bureau of Labor Statistics, the O*NET database from the US Department of Labor, and its ain LinkedIn Occupational Graph.

This data, on with information gathered from Agent 365, is aggregated into investigation intended to answer:

  • How overmuch bash idiosyncratic agents, models, and users cost?
  • What concern activities is AI performing (across much than 6,000 defined usage cases)?
  • How overmuch quality enactment would those activities would ordinarily require?
  • What is that enactment worthy successful dollar terms?
  • Does expanding oregon reducing AI spending makes concern sense?

The extremity is to supply a broad representation of AI walk against worth produced.

Measuring AI arsenic a labor/software hybrid

Software accounting is agelong established and straightforward. We measurement subscription fees, outgo per seat, compute costs, load costs (like level abstraction and different infrastructure), and more. But AI enactment doesn’t acceptable that model, making it precise hard to quantify value.

Instead, Agent 365 Business Analytics measures outgo some comparative to quality labour and successful presumption of technological expense. For example, it mightiness look astatine a task and find however overmuch quality effort that task would usually take, and past terms that effort based connected prevailing communal labour rates.

Then, it mightiness origin successful however overmuch of the task was really done by a quality vs. however overmuch was contributed by an AI, on with the estimated worth of the AI’s contribution. Then it mixes successful the existent AI processing outgo (whether successful subscriptions, tokens, on-premises infrastructure, and unreality support).

Also: These Claude Enterprise features fto you way AI ROI – what’s new

What comes retired is simply a hybrid labour estimation that combines quality contribution, human-replacement savings, AI costs, and past perchance AI savings implicit quality costs. It sounds positively Orwellian. But the bottommost enactment is that companies truly haven’t been capable to recognize the AI bottommost line.

Ah, but estimated worth isn’t realized savings

There’s a catch. All these numbers are estimates. In its blog post, Redmond specifically says, “Assisted hours are a directional estimation of enactment AI helped complete, not measured clip saved oregon fiscal return.”

In different words, what Agent 365 Business Analytics estimates mightiness not really enactment up with real-world acquisition oregon expenses. It’s besides imaginable that AI enactment mightiness person been assigned due to the fact that determination was AI to bash the work, not that the AI was redeeming institution labour costs with each token processed.

In different words, Microsoft whitethorn beryllium capable to archer you the economical worth of AI-assisted work, but it can’t show definitively that the institution saved immoderate existent money.

Also: 5 reasons your institution needs a main AI serviceman successful 2026

I find it peculiarly telling that this is from the institution that makes PowerPoint. Because now, with Agent 365 Business Analytics, it’s wholly imaginable that you could nutrient economical dashboards that look really, really impressive, but are wholly irrelevant to existent life.

That’s not to accidental the merchandise is useless. But it does spell to the thought that you volition person to beryllium rather disciplined successful however you construe the results and however overmuch banal you enactment successful them for decision-making. It is one root of imaginable truth, not the root of implicit truth.

Microsoft wants to beryllium the power halfway for AI

Microsoft has been investing for decades to physique retired the halfway infrastructure of endeavor computing. AI isn’t immoderate antithetic erstwhile it comes to the company’s wide mission.

With Agent 365 and the company’s related offerings, Microsoft wants to assistance you:

  • Monitor AI agents from aggregate vendors.
  • Manage their identities, permissions, and security.
  • Track and bounds AI spending.
  • Evaluate whether agents are delivering value.
  • Decide which AI systems merit further investment.

This means the institution doesn’t needfully person to supply you with the AI exemplary to assistance you tally your enterprise. With integration with offerings from OpenAI, Anthropic, Salesforce, Google, Databricks, Snowflake, and others, the institution doesn’t adjacent person to supply the apps oregon SaaS capabilities.

But if it tin get successful determination and negociate your plumbing, assistance you power costs, and beryllium the HR cognition for your AI workforce, Microsoft could ain a cardinal portion of your business’s operations.


You tin travel my day-to-day task updates connected societal media. Be definite to subscribe to my play update newsletter, and travel maine connected Twitter/X astatine @DavidGewirtz, connected Facebook astatine Facebook.com/DavidGewirtz, connected Instagram astatine Instagram.com/DavidGewirtz, connected Bluesky astatine @DavidGewirtz.com, and connected YouTube astatine YouTube.com/DavidGewirtzTV.

David Gewirtz

David Gewirtz

Senior Contributing Editor

Computer subject prof turned AI innovator David Gewirtz has spent implicit 30 years driving advancements astatine the forefront of artificial quality and is the recipient of the Sigma Xi Research Award successful Engineering. He wrote a pioneering investigation that helped signifier the aboriginal speech astir AI ethics. He was besides a pioneer successful the commercializing of AI products, including AI languages and knowledge-based systems. He is the decorator of the Al Editor, an experimental artificial quality motor for parsing, classifying, and identifying quality stories. David is simply a subordinate of the Association for the Advancement of Artificial Intelligence. He presently serves connected the Artificial Intelligence Threat and Mitigation Cross-Sector Council (AI CSC) for InfraGard, a concern betwixt the FBI and manufacture leaders for the extortion of U.S. Critical Infrastructure. He is the writer of Where Have All the Emails Gone?, arsenic good arsenic How To Save Jobs and The Flexible Enterprise.

Read Entire Article