Responsible AI
Systems that act deserve systems of accountability.
As software takes on more of the work, the standards we hold it to matter more than the capability it demonstrates.
Responsible AI
Trust is a design constraint, not a policy page.
Privacy by Design
Collect only what is necessary.
User Agency
Humans retain meaningful control.
Security
Sensitive information deserves exceptional protection.
Transparency
Important AI-generated decisions should be understandable and traceable where appropriate.
Evaluation
Systems should be continuously tested against real-world outcomes and failure modes.
Human Benefit
We build AI to increase human capability and agency.
These are the principles we hold ourselves to. We do not claim certifications, external audits, regulatory approvals, or specific security capabilities that have not been independently verified. Where a portfolio company achieves such verification, it will be stated explicitly on that company's page.
The future isn't something
we wait for.
We build it.