k
keynews.ai
keynews.ai
WorkThe story, in brief

If agents are building your app, who gets the W-2?

Autonomous agents are writing production code. Now accountants are asking: should AI labor count as a salaried employee under GAAP?

Reporting by Vercel Blog 1 min brief
Illustration of independent geometric mechanisms passing paper tasks along branching amber tracks.
AI agents and the coordination of work.AI illustration by KeyNews
The KeyNews take

Why it matters

As AI agents move from capability demos to production workflows, the legal and accounting frameworks that govern software development are breaking down. This isn't just philosophy—it's a real question about how to capitalize (or expense) AI-generated work, with direct implications for SaaS margins, valuation, and workforce economics.

The key facts

5 to know

  1. Autonomous agents now design, build, test, and deploy full-stack features end-to-end

  2. Current GAAP allows capitalization of third-party software costs only if directly tied to software creation

  3. Historically, developer tools treated as overhead because cost couldn't be tied to specific deliverables

  4. When agents perform work that meets capitalization criteria, they should be treated similarly to salaried engineers under current accounting rules

  5. Published by Vercel on Aug 18, 2025

Go to the source

Vercel Blogvercel.com

Publisher excerpt: Autonomous coding agents are not the future. They are already . Agents can now design, build, test, and deploy an entire full-stack feature from front end to back end without a human touching the keyboard. here The reality is that while this technology has advanced quickly, Generally Accepted…
Read original report
Back to today's editionMore work news

The wider picture

View all
Illustration of a transparent lens revealing connected networks across layers of paper.
AI illustration by KeyNews
Work01

The next AI advantage isn’t a bigger model — it’s a better semantic layer

As agentic AI moves from pilots to production, semantic alignment—shared, machine-readable business definitions—has become a prerequisite for safe deployment. This is a people and process problem, not a model problem, and it requires organizational work that most data teams aren't yet structured to do.

CIO
Illustration of two anonymous hands arranging task cards around an amber tool on a shared desk.
AI illustration by KeyNews
Work02

I Built AI Clones of My Coworkers. Things Got Weird

As AI personalization deepens in workplace tools, the line between mimicry and utility blurs. This is a cautionary tale about how AI-as-coworker products could amplify personality performance over actual work.

Wired AI
Illustration of two anonymous hands arranging task cards around an amber tool on a shared desk.
AI illustration by KeyNews
Work03

‘Rising threats and under-resourcing for cybersecurity is taking a toll on the people tasked with managing it’: Cyber teams are being pushed to breaking point – and AI is doing little to alleviate strain

AI is being marketed as a cybersecurity force multiplier, but practitioners report it's not reducing workload or burnout in under-resourced teams. This is a workforce reality check on AI's promised impact on a critical profession.

ITPro