AI Ethics Policy

Updated 11 September 2018

AI should be ethical.
Just because AI can do amazing things, doesn't always mean it should. We believe companies which employ AI should commit to what they will and won't use it for, and be transparent about this.

  1. Our AI will not analyse psychological vulnerabilities for ad targeting
  2. Our AI will not deliberately create ads that mislead or manipulate people
  3. We will make every effort to minimise biases inherent in AI training

AI should be sustainable.
AI shouldn't just increase productivity and profit for companies - it should increase quality of life for people, too. In a world where AI can increasingly do things people can, we believe companies have a collective responsibility to ensure those people aren't disenfranchised by it. That's sustainable AI.

  1. If our AI automates part of someone's job, that person should benefit
  2. Use of our AI should elevate the people interacting with it
  3. The decisions our AI makes must be supervised, interpretable or both

This policy is an early attempt to set good ground rules in an unregulated industry. We will update the policy from time to time as the technology, and its use cases, evolve.

Contacting Us
If there are any questions regarding this document, you may contact us using the information below.

Pencil AI Limited

151 Rosebery Avenue London EC1R 4AB