In short: Google quietly launched an API that allows its Gemini model to ingest live streams, capturing what’s happening on your screen (or via a video feed), along with audio or text, and to respond instantly as you work. Most teams saw this as a cool tech demo and moved on.

At PAIR, we saw the missing piece in building truly AI-native workforces.

Within hours we were white-boarding. Within days we'd built a prototype. And a couple of months later, we're introducing something that shouldn't exist yet: objective proof of AI capability that goes beyond self-reporting and guesswork.

That's how fast the world moves when breakthrough tech meets a team obsessed with turning AI adoption into measurable business impact.

The adoption gap that's costing millions

Here’s what we see every day helping organisations build AI-native workforces: the gap between AI access and AI adoption is enormous.

Companies invest millions in Copilot rollouts and ChatGPT Enterprise licences. Employees get excited, try it a few times, then gradually drift back to old ways of working. Six months later, usage analytics tell a depressing story: low active users, minimal productivity gains, questionable ROI.

The problem isn't the technology. It's that real AI capability - the kind that transforms how work gets done - requires deliberate skill development. And until now, there’s been no objective way to measure where people actually stand.

Building capability, not just awareness

At PAIR, we spend our time perfecting AI upskilling that creates certified, measurable outcomes. Our platform takes professionals through progressive levels (Fundamentals, Proficiency, Mastery) with hands-on tasks tailored to their actual work on a person-by-person level.

But we kept hitting the same friction point: people arriving with wildly different baseline skills. Some overestimate their abilities, others underestimate. Traditional pre-assessments rely on self-reporting or multiple choice questions that bear no resemblance to real AI usage.

So when Google's breakthrough API landed, we saw an opportunity to solve assessment the same way we solve learning: through real work, not artificial tests.

Proving capability through practice

Meet Maia, our new AI assessor that fundamentally changes how we measure AI fluency.

Instead of asking people to rate their prompt-writing skills on a scale of 1-10, Maia gives them a real task and watches them work for ten minutes.

Share your AI workspace tab, work naturally with your preferred tools, and Maia scores your actual technique: prompt structure, iteration strategy, output verification, efficiency.

The feedback reads like notes from an expert coach giving you a clear picture of where you shine, and where there is room for improvement.

The breakthrough moment

Early users describe something unexpected: the ten-minute session changes how they think about AI permanently.

When you know you’re being evaluated on prompt clarity, you start writing clearer prompts. When effectiveness matters, you stop accepting mediocre first outputs.

For some, it's validation: confirmation they’ve genuinely developed advanced skills.
For others, it’s revelation: discovering specific gaps they didn’t know existed, like weak prompt constraints or skipping output verification.

But for everyone, it eliminates the guesswork that’s been holding back AI adoption.

From individuals to AI-native teams

Maia is just one part of how PAIR builds AI-native workforces. The real transformation happens when skills assessments lead to action.

If you demonstrate foundational capability, you can immediately move on to more advanced challenges, building AI agents, streamlining research workflows, or applying AI to real business problems.

If gaps are revealed, you’re directed to a hands-on fundamentals course that builds practical skills in prompting, iteration, and applied research. No wasted time on content you’ve already mastered. No skipped steps that undermine long-term gains.

This precision (getting the right people on the right path at the right time) is what unlocks faster progression and real productivity lift.

The result? Organisations finally see the ROI they expected from their AI investment, grounded in certified capability they can measure and improve over time.

Our team moves quickly - CEO James and Developer Adrian talk about the launch of this new feature

The ten-minute test that changes everything

Maia isn’t about surveillance or exams. She’s about clarity.

• Clarity for individuals: where you’re strong, where you can grow, and how to get there.

• Clarity for leaders: where real capability sits across teams, and where targeted support is still needed.

That clarity is what turns AI potential into performance. And it starts with just ten minutes of real work.

Ready to see where you stand? Meet Maia on pairnow.ai!