Think· Criterio
We don't teach tools: we build organizational judgment.
Training by profile so that the use of AI is safe, useful and aligned with real business decisions. The result is an organization that knows when to use AI, when not to, and why.
Criterion, from the Greek krínein: to separate, to judge.
Why it matters
Coordination without central control only works if every part has judgment.
Adopting AI means getting the organization to coordinate by deciding with its own judgment.
It changes the cost of work, affects the whole P&L and redesigns expert work. And it demands in-house capability by role, an AI committee with a mandate, and risk management.
Judgment doesn't live in isolation: ambition and priorities are set in Guide, knowledge is structured in Ground and execution is validated in Accelerate.
Who it's for
Designed for those who build the organization's judgment
- HR and Transformation leaders driving AI adoption.
- CEOs who want to build an AI culture, not just buy tools.
- Executive committees that need the organization ready for what's coming.
What we see
Why does AI training fail?
They teach the tool, not judgment
tool-itis
People are trained on the product of the moment and taught to “write prompts” instead of building judgment that lasts. The tool changes, and the training has expired.
They train without touching real decisions, and without practice
The training inspires, but that inspiration evaporates on the way out of the room. There's no follow-through. Anything that isn't tied to a real decision in the job changes no behaviour.
One size for everyone
one-size-fits-all
The CEO, middle managers and the whole organization get the same content. Without tailoring to each role, it's too much for those at the top and too little for those below.
They train the person without changing the environment
Individual judgment is handed to an organization with no governance and no semantic layer. Knowledge hits friction and never becomes capability.
They measure attendance, not decisions
vanity metrics
They celebrate how many people went through the training instead of what is now decided better. Training piles up without moving the needle.
Five audiences, six capabilities
Each profile exercises judgment in a different way
Judgment is the ability to make good decisions about AI. We build it according to how each profile relates to AI: using, structuring, deploying, governing or leading.
-
They use AI with judgment and autonomy.
- CognitiveStructuring what you ask AI and judging what it returns: framing well, spotting answers that don't hold up and deciding when not to trust it.
- CollaborativeMaking knowledge about AI flow between people and teams, so that what one person learns speeds up adoption for everyone.
-
They build the semantic layer that gives AI its meaning.
- LogicalDefining how AI decides within a process: its scope, its limits and the interaction between person and machine.
- OperationalTaking AI from pilot to production and keeping it reliable: drift control, timely intervention and continuity once it runs on the business.
- CollaborativeMaking knowledge about AI flow between people and teams, so that what one person learns speeds up adoption for everyone.
-
They oversee and tune AI, and make it scale.
- OperationalTaking AI from pilot to production and keeping it reliable: drift control, timely intervention and continuity once it runs on the business.
- CollaborativeMaking knowledge about AI flow between people and teams, so that what one person learns speeds up adoption for everyone.
-
They decide strategy, risk appetite and investment.
- OrganizationalEstablishing who owns each decision and who is accountable for it, and transferring expert knowledge before it's lost.
-
They turn strategy into capability, team and decisions.
- AdaptiveDesigning adoption so you can switch model or vendor without rebuilding operations, and leading when the context moves faster than the structure.
How we train
Our training sessions aren't classes
They are workshops that shape the strategy, co-creation that rolls out governance, sensors that design internal change, and labs that activate the organization.
AI activation
Catalyst network
Professionals in your organization who become internal agents of the transformation: they find use cases in their areas, connect knowledge across teams and sustain the capability once we're gone.
People who know their area of the business inside out, with natural influence in their teams and curiosity about AI. They don't need to be technical.
As in chemistry, a catalyst…
- speeds up the transformation without being consumed.
- remains after the reaction.
- makes possible changes that wouldn't happen on their own.
- works for many transformations.
AI Act, Article 4
AI literacy is now an obligation
Article 4 of the AI Act, in force since 2 February 2025, requires organizations to ensure their staff have a sufficient level of AI literacy, according to their profile and how they use it.
What your organization must be able to show
- Documented training by role and level.
- Tailored to actual use.
- Renewed when tools change.
- With attendance and materials on file.
In Spain, the Council of Ministers approved the Bill for the good use and governance of AI on 26 May 2026.
How we develop it
What we draw on to develop Think
Tap each point to see what it involves.
AI literacy
Paths by profile, safe-use guides and a generative AI policy. Types of AI (predictive, generative and agents), good and bad uses, real risks and the judgment to decide, with practice on cases tailored to the sector and the role.
Executive education
Programmes for boards and senior management: judgment to design and decide on AI, not technology outreach. AI strategy and governance workshops and hands-on use-case workshops for executives.
Executive coaching for CAIOs
Support for a technical profile (CTO) to make the leap to Chief Data & AI Officer, and leadership development to govern a cross-cutting discipline like AI.
Catalysts and change management
Co-design with teams, an internal Catalyst network and adoption metrics. AI knowledge workshops, use-case discovery and prioritization, and pilot building.
Knowledge engineers
Training to build the semantic layer that gives AI the meaning of the business: from glossary to ontology, elicitation, governance of meaning and operationalization. Participants finish with a minimum semantic layer on a real domain.
Let's talk about building judgment in your organization.
You can start with a Judgment Clinic: bring a real case and leave with a decision map, a delegation checklist and a prototype.