Presentation Notes: Strategizing Artificial Intelligence and Machine Learning (Kelly Calhoun Williams)

• Need to leverage technology to scale the education process (e.g. AI to assess writing samples)
• In education environment, it is important to have a sense of the quality & quantity of data and assess level of confidence. Automation is required for volume
• Researchers are clamoring for AI because of the volumes of data they have to interpret
• To place AI as strategy, you need to start at the foundation, which is data. There is no value in iterating on bad data.
• Hype cycle: digital ethics very low in the hype cycle (and not in education) - deep and machine learning at the peek of the hype
• Higher ed: Virtual Assistants
• All human data is biased which impact the algorithms - Algorithmic Bias
• Algorithmic aversion
• Success: narrow focus, goldilocks principle with data, have the payout, build talent and skills

Questions

• Digital ethics questions - law suites connected to AI, biases, ..., what is our responsibility?
  • Higher ed due diligence - cannot guarantee risks exists: have system in place what to do when something goes wrong