KM and AI, Working to Eliminate Bias and AI’s Role in the Future
The news these days can seem grim with reports that robots will eventually lead to mass unemployment. However, there’s a human – computer digital divide, Daniel W. Rasmus, founder & principal analyst, serious insights and Bellevue College, Faith and AI, explained during his session, “Rethinking KM For An Age Of AI & IoT” at KMWorld 2018.
The 22nd KMWorld conference is co-located with Taxonomy Boot Camp 2018, Enterprise Search & Discovery, Text Analytics Forum ’18, and Office 365 Symposium.
As machines get better and better at sensing the environment, comprehending content, interpreting signals, and anticipating needs, there will always be a need for a human element, Rasmus said.
“I don’t think we will have fully conscious machines,” Rasmus said.
The underbelly of AI is incomplete, Rasmus explained, and there is a bias hidden behind recommendations and suggestions that AI and IoT machines pick up.
“Human intent of the system is key,” Rasmus said. “Not a lot of AI is making decisions on its own and when we call on autonomous technologies it has a difficult time on its own.”
He emphasized that knowledge managers need to create balance between the depth and history of infrastructure and the moment, which comprises of insight and data.
He recommended:
- Design your AI and IoT KM experience:
- Team KM and AI/ML for:
- Data design
- Test for bias
- Determine what questions to ask
- Look for shifts in underlying assumptions (when does the model stop working)
- Team KM and IoT for:
- Environment design
- System design ad integration of existing data
- Leveraging insights to find vale
“Become your organizations conscience,” Rasmus said.
Many speakers, including Rasmus, have made their presentations available at www.kmworld.com/Conference/2018/Presentations.aspx.
KM World 2019 will be held November 5-7, 2019 at the JW Marriott in Washington, DC, with pre-conference workshops on November 4.