AI Trust and Safety Principles - take a look!

[Alies Maybee]

This past April we had the Vector Institute present a webinar to us about Artificial Intelligence (AI) in healthcare and had an enthusiastic response.

They have since posted their AI Trust and Safety Principles. Take a look and let me know what you think. Have they covered it? Should more be added, modified, taken away?

In summary, here are their 6 principles:

• AI should benefit humans and the planet

• AI systems should be designed to reflect democratic values

• AI systems must reflect the privacy and security interests of individuals

• AI systems should remain robust, secure, and safe throughout their life cycles

• AI system oversight should include responsible disclosure

• Organizations should be accountable

If there is interest, we can arrange another conversation with them about these principles.