Cognitive Commons An independent research initiative

How much do we really understand about the human intelligence AI is replacing?

Cognitive Commons studies the judgement, know-how and everyday coordination that keep organisations running as AI is adopted, and what happens when it is automated before anyone has understood it.

Add your organisation to the picture.

Take the 5-minute survey
  • Five minutes
  • Confidential
  • You get the benchmark report
What is Cognitive Commons

A commons for what organisations know about people and AI.

A commons is something held in common: a shared resource that many people contribute to, and that everyone who contributes can draw on.

That is the idea here. Cognitive Commons pools what organisations are learning about the human side of AI, across roles, sectors and industries, into one shared body of knowledge. You contribute your experience through a short survey, and in return you get the collective picture back. Nothing is taken and locked away. The point of a commons is that it gives back to the people who build it.

Why this matters

AI is not creating the problem. It is revealing it.

Organisations are adopting AI faster than they understand the human work it touches. Most cannot see, describe or measure the tacit human intelligence their work depends on, so they automate before understanding the value they are removing. Productivity is not the same as value.

81%

of companies fail to turn AI into real business value.

Source: DLA Piper
95%

of enterprise GenAI pilots deliver no measurable impact on profit or loss.

Source: MIT Project NANDA, 2025
Who we want to hear from

This is about your whole organisation, not one function.

The human side of AI is felt everywhere decisions are made and work gets done. We want to hear from all of it.

01

Transformation & change

You own the operating model, not just the technology.

02

Operations & COOs

You see why pilots stall long before they scale.

03

People & HR leaders

You know where capability really lives, and what is at risk.

04

AI & data leaders

You live in the gap between a working pilot and value at scale.

05

Risk, governance & compliance

You have to evidence oversight someone else designed.

06

Security & CISOs

You inherit AI decisions you were never part of.

What we’re investigating

The questions the survey explores

  1. Whether AI strips out human know-how that no one had captured.
  2. Why so many AI initiatives stall short of the value expected.
  3. How much say you have in AI decisions, versus how much you are accountable for making them work.
  4. Whether a “human in the loop” is genuinely designed, or merely asserted.
  5. How deliberately organisations map the human judgement their work depends on before automating it.

These are fast becoming legal questions, too. The EU AI Act and UK data and equality law increasingly turn on evidencing a documented human in the loop, owned by the organisation deploying the AI, not the vendor.

And in security, unsanctioned “shadow AI” is now found running in around 75% of organisations (2026 CISO AI Risk Report). Less a threat to stamp out than a signal of where the official process is failing people.

Taking part

Two ways in. Both start with five minutes.

Step one

Take the 5-minute survey

Open to anyone, fully confidential, and reported only in aggregate. You will add a data point, see where your organisation stands, and receive the benchmark report.

Step two, optional

Go deeper, if you’d like

A confidential 45-minute conversation, opted into at the end of the survey. This is where the real understanding comes from, and it is entirely up to you.

Take the 5-minute survey
What you get back

A benchmark, not a black hole.

Too many surveys take and never give. This one gives back. Every contributor receives the benchmark report: how organisations like yours are really handling the human side of AI, where you sit against your peers, and what the leading edge looks like.

Everything is held under the Chatham House Rule. Findings are shared; contributors are never named.

Who’s behind it

An independent initiative from Invistra.

Cognitive Commons is a research initiative of Invistra Ltd, an independent, vendor-neutral firm working on how organisations design for human contribution in AI-enabled systems.

The research is not sponsored by any technology vendor, and it is not selling anything. Its purpose is to build a shared, honest picture of a problem most organisations are facing largely alone.

Add your organisation to the picture.

Take the 5-minute survey
Five minutes · Confidential · You get the benchmark report