Our logo represents a goat.
And that’s no coincidence.
Goats are intelligent, tenacious, persevering animals that can climb peaks that others cannot even imagine. Just like us.
We are always ready to find the right path. The most original, most effective, most unconventional one.
We use AI not to follow the herd, but to blaze new trails. Because true power lies in having a clear destination and the agility and ability to guide our partners.
OURMANIFESTO
Unconventional by nature.
Responsible by choice.
Artificial intelligence can do anything. But it does nothing without direction.
We chart that direction with strategies designed by people. For the people.
This is how we build our projects: with a radically human approach that puts the customer and their specific needs at the centre.
An approach we call “augmented humanity”.
A way of working that doesn’t just automate, but unleashes energy, opens up perspectives, and helps ideas come to life – strategic, creative, inclusive.
Because for us, AI doesn’t replace. It amplifies.
It amplifies insights. Data. Questions. Room for improvement.
It also amplifies our responsibility: towards people, towards the planet.
Every solution we build takes into account the impact it can generate.
We want to be free to think, choose and create.
Always respecting the ecosystem – social, human and environmental – in which we live.
We are not interested in automating to do more.
We are interested in humanising to do better. And in a fairer way.
That’s why we are more than an agency: we are a digital partner that thinks, listens and creates.
ETHICAL CODE
Human driven,
securely powered
For Keyformat, innovation is not a race without rules, but a conscious journey. AI does not replace our talent, it amplifies it. But for this to happen, discipline is needed.
We have transformed our internal policies into a transparency pact with our partners. Here’s how we manage AI to ensure safety, quality, and respect.
1
No to public data
We never enter confidential, strategic, or personal information into public tools that use data for training purposes.
2
Yes to protected environments
We only use suites with binding data protection agreements. What we process remains in a closed, secure circuit that complies with the GDPR.
3
No autopilot
AI is our research and brainstorming assistant, but the final decision is always made by humans.
4
Fact checking
We check sources, data, and statements to ensure the integrity of information, applying a rigorous human filter.
5
We test safely
We test new developments in a controlled sandbox environment to innovate without risk, using only approved and secure tools for our customers.
6
Copyright compliance
We don’t use AI to circumvent copyright. We create original content, where technology is the means, not the end.
7
Monitoring biases
We constantly monitor outputs to avoid biases or discrimination that models may inherit.
8