Apple Researchers Built a Virtual User to Test AI Assistants Before Real People Do
A new research framework simulates the back-and-forth of real app use, so proactive AI assistants can be tested and scored without putting actual users at risk.
Key points
- Apple ML Research published PARE, a framework for testing AI assistants that act without being asked, in 2025.
- PARE models apps as finite state machines, a structured map of every screen and action an app can offer, instead of simple command lists.
- Current AI assistant testing methods miss the step-by-step, stateful nature of real app use.
- The framework makes it possible to simulate a realistic user so developers can measure how well a proactive agent performs before it touches a live product.
Most AI assistants wait for you to ask them something. A proactive agent, software that watches what you are doing and acts on your behalf before you even type a request, is a different animal. It might notice your flight is delayed and rebook it, or spot a bill due tomorrow and schedule the payment. Useful, yes. Also risky if it gets things wrong.
The problem is that building and testing these assistants is hard. You need a realistic stand-in for a human user, someone (or something) that clicks through menus, fills in forms, and changes their mind mid-task, the way real people do. Until now, no good tool existed for that.
Apple ML Research introduced PARE, the Proactive Agent Research Environment, to fill that gap.
What does this actually change for developers?
It gives them a safe sandbox to break things before shipping. PARE works by mapping each app as a finite state machine, think of it as a detailed flowchart showing every screen, every button, and every path a user can take through the software. That is a much richer picture than the flat lists of commands that earlier testing tools relied on.
With that map in place, PARE can run a simulated user through realistic sequences: open the calendar, check Tuesday, see a conflict, try to reschedule. The AI assistant being tested has to respond to each step, just as it would with a real person sitting at the keyboard.
Previous frameworks treated apps like a vending machine: put in a command, get an output. PARE treats them like a conversation that unfolds over time, which is far closer to how people actually use software.
What does this mean for ordinary people?
Nothing changes in your apps today. This is research infrastructure, the behind-the-scenes scaffolding that AI teams use to check their work.
But the downstream effect matters. Proactive agents are coming to phones, laptops, and productivity tools faster than most people realise. A testing framework like PARE means the teams building those agents have a way to measure whether the assistant helps or causes chaos before it lands on your device. Think of it as a crash-test dummy for AI software.
One honest caveat: a simulated user is still a simulation. Real people do unpredictable things, and no lab environment captures all of that. Good test results in PARE are a necessary step, not a guarantee that a proactive agent will behave perfectly in the wild.
Takeaway: If you use productivity software and wonder whether AI assistants will one day manage your calendar or inbox without asking, the answer is yes. Frameworks like PARE are how researchers check that those assistants will not make a mess of it first.
