About ENTROPY
ENTROPY builds the layer between AI agents and the systems they are allowed to touch. It was founded in Berlin by three people working across machine-learning research, security research and defence cybersecurity.
We are in private development with a small number of design partners whose agents work with sensitive data.
Founders
| Name | Role | Background |
|---|---|---|
| Natalie Ma | Chief Executive Officer | Machine-learning research at Stanford. LBS startup finalist. LinkedIn |
| Ivan Klisanin | Chief Technology Officer | Security researcher. Has founded two companies. |
| Oscar Grillborzer | Chief Product Officer | Cybersecurity for the German military. |
Principles
- 1
Least authority. An agent should hold exactly what its task needs, for as long as the task lasts.
- 2
Evidence over opinion. Every decision is recorded so that it can be reconstructed and replayed.
- 3
Revocable by default. Any authority that can be granted can be withdrawn at once.
- 4
Quiet by design. Invisible while agents behave. Decisive when they do not.
Affiliations
- Oxford University Innovation, Incubator Phase 1
- CODE University of Applied Sciences, Berlin
Working with us
If you work on agent security, authorization or provenance, write to one of us. Contact details