Trend Alert! Updating Your Code to Account for Artificial Intelligence

I was at the European Compliance and Ethics conference last week when a new trend came across my radar. In the first session, John Bowman, AI Ethics Market Strategy Lead for IBM, began talking about how IBM has approached using artificial intelligence (AI) within their products. He said that they had updated their Code of Conduct to account for AI-related concerns.

Within days, two of Spark Compliance Consulting’s other clients told me they were updating their Codes for the same reason. Ladies and gentlemen, updating your Code to account for AI is officially a new trend.

Updating your Code to account for AI

The Purpose of the Code

The Code of Conduct is meant to be a resource for employees to learn, at a high level, about what doing the right thing means at your company. It can’t account for every issue, but it should lay out the principles upon which people should make decisions.

With that in mind, surely the Code doesn’t need to be updated just to speak to one issue, right? Behavior should be guided by integrity in all circumstances, not just those specifically laid out in the Code. However, as employees and employers fixate on the use of AI, both in their products and by their employees, focusing a part of the Code on AI becomes worthwhile.

Ethics by Design

Mr. Bowman talked about the inclusion of “ethics by design” as part of IBM’s strategy in creating and utilizing AI-related products.

The idea of “ethics by design” comes from the requirement in the European General Data Protection Regulation that calls for “privacy by design.” Privacy by design means that concerns about data privacy are baked into the thought process from the very beginning of a project, and that privacy controls are considered throughout the product design and marketing.

“Ethics by design” is my new favorite phrase, especially when it comes to managing AI risks. Mr. Bowman talked about how ethics by design means that IBM considers the problems of bias when it begins a project including AI. It thinks about guidelines and controls, along with the initial information fed to the AI engine at the beginning and throughout the project.

What to Update

A new section of your Code relating to AI doesn’t have to be long or overly detailed. It should:

  • Recognize the risk of AI’s harm, including bias or misinformation
  • Reference the company’s values and describe how ethical use aligns with them
  • Point to the ethical decision-making framework within your Code if you have one
  • Tell people to call the Compliance Department for answers to questions, and include contact information on the page
  • Reference the section on speaking up/hotline if people want to report concerns
  • Point to a larger AI-related policy, if your company has one
  • Remind people to use ethics by design in creating AI-based products (if applicable).

Celebrate the Changes

We as compliance officers recognize that ethical behavior extends to use of AI whether the Code specifically calls it out or not, but the average employee may be unable to make the leap without a clear statement.

Updating your Code and celebrating the changes gives you an opportunity to communicate to the company about the Code more generally. It also showcases that Compliance is up-to-date and following the most important emerging issues.

Take the opportunity to follow the trend. This one’s a good one!

Want help updating your Code of Conduct? We’re here for you!

See www.sparkcompliance.com or contact me at kgranthart@sparkcompliance.com for more details.

Share the blog!

Twitter
LinkedIn
WhatsApp
Email
Print
Picture of Kristy Grant-Hart

Kristy Grant-Hart

Kristy Grant-Hart is the founder and CEO of Spark Compliance.
She's a renowned expert at transforming compliance departments into in-demand business assets.