As AI agents grow ever more capable, enterprises racing to put them to work across applications, workflows, and products face a new challenge: ensuring an agent does what it’s supposed to do when it’s deployed across different environments.

Microsoft is trying to solve this problem with a new open source standard called Agent Control Specification (ACS) that aims to give developers a more consistent and granular way to control what AI agents are allowed to do.

The specification essentially lets developer, compliance, and security teams define their own policies for agents to follow. The rules can define what the agent may do, what it must not do, when a human should approve an action, and what evidence should be logged for later review. These policy files are checked at several “interception points” when the agent is off performing a task to make sure it stays within the guardrails.

The spec comes as developers are improvising ways to control what their AI sees and does, especially with conversations focusing on AI workflows going wrong due to tool misuse, or unintended actions that result in cascading failures.

Today, developers might specify instructions in a system prompt, add custom checks in the application code, or use classifiers to catch problematic inputs and outputs. Those approaches work, but they often leave companies with fragmented controls that are hard to audit and harder to reuse across different frameworks, interfaces, and systems.

ACS aims to integrate those controls into a common governance layer. Microsoft says the specification can be used to check whether an agent is sticking to guardrails at multiple points in its workflow — before it receives input, before it calls a tool, after a tool returns a result, and before the final response is sent to the user. A policy may allow an action, block it, redact sensitive information, or even ask a person to approve it.

Developers can also insert classifiers for inputs and outputs to categorize information, predict outcomes, or determine how an agent should respond; add LLMs with prompts to act as a “judge” for policies; and logic for checking tool calls, tool selection, input accuracy, output usage, and responses.

And because these policies can be written as single files, they can be bundled with agents, allowing a security policy to follow an agent across different frameworks and environments.

ACS is shipping as an SDK with plug-ins for LangChain, the OpenAI Agents SDK, the Anthropic Agents SDK, AutoGen, CrewAI, Semantic Kernel, Microsoft.Extensions.AI, MCP tools, and more.

Nvidia opened Taipei’s enormous Computex trade show on Sunday with a spark, literally. The chipmaker unveiled a new PC CPU called the RTX Spark, which it dubbed a “superchip,” and named a who’s who list of PC makers that will soon deliver AI PCs powered by it.

The super-fast, 1-petaflop chip is designed to run AI agents like OpenClaw or Hermes Agent securely, according to Nvidia. Such RTX Spark Windows PCs will be available this fall from ASUS, Dell, HP, Lenovo, Microsoft Surface, and MSI, with models from Acer and Gigabyte to follow.

In addition to being equipped with secure sandboxes (jointly developed with Microsoft) to run agents securely, the PCs will also have enough CPU, GPU, RAM, and underlying Nvidia CUDA software to run local versions of large language models.

Nvidia said that its RTX technology will deliver faster performance for AI, better image quality, and support for AI features in more than 1,000 games and applications.

The chipmaker is marketing this as an alternative for creators making AI content, as well as providing a significant upgrade to its traditional market of gamers. Nvidia said more than 100 Windows software makers have signed on to support the new chip, including Adobe, Blender, ComfyUI, Riot Games, and Xbox.

But Nvidia founder and CEO Jensen Huang’s vision for these new PCs is far larger. He wants to end the days of launching apps, pointing, clicking, and typing.

“With RTX Spark and Microsoft Windows, you ask — and the PC does the work,” he said in the press release. “Frontier models. Creative workflows. RTX games. All on a laptop.”

Last month, after delivering another record quarter, Huang promised investors he had found a new $200 billion market for Nvidia in selling CPUs for AI, not just GPUs. He made specific mention of the high-end server CPU released earlier this year called Vera — of which Nvidia says it has already sold $20 billion worth.

He also hinted at his bigger ambitions. “We’ll have billions of agents, and those billions of agents will all use tools. And those tools are going to be like PCs, just like us humans using PCs today,” he said on the earnings call in May. “We’re going to need a lot more CPUs.”

Nvidia ARM-based Windows devices have been tried before — and failed. Back in 2013, Microsoft famously had to write off $900 million on its Nvidia ARM-based Surface RT, with partners like Dell also bailing on the product.

But at this point, after delivering record after record of quarterly revenue, it’s hard to bet against Huang as he pursues his PC dreams once again.

And this chip is an entirely different beast. It’s more powerful, not less. Microsoft is positioning its own RTX Spark PC as so mighty that it named it the Surface Laptop Ultra, and is calling it “the most powerful Surface Laptop ever built.”

Still, PC manufacturers have not released a lot of specifics about each of their offerings, including pricing. These systems appear to be full-fledged Windows versions of the DGX Spark mini-computer that Nvidia already sells to developers for about $4,800.

We’ll have to wait and see if these PCs will compete on price with the affordable Mac Mini that has become a popular choice for running OpenClaw. Or perhaps they will sit at the high end of the PC market, like Nvidia’s own agent-running mini computer.

Either way, if Nvidia has cracked the code on bringing AI agents easily, safely, and usefully to the masses, it could — and should — be big.