Skip to content

Latest commit

 

History

History
40 lines (25 loc) · 2.21 KB

townhall.md

File metadata and controls

40 lines (25 loc) · 2.21 KB

Townhall

Over a period of several weeks, a group of technologists took an AI townhall event on the road, visiting Bergen, Oslo, Stavanger, Trondheim, Stjoerdal, and Harstad and reaching more than 3000 employees.

Team

  • A leader in corporate communications with a special interest in tech and AI
  • A scientific programmer
  • A data scientist specializing in NLP
  • The Leading Advisor in data science
  • A data scientist specializing in responsible AI

Content

The presentation contains:

  • Welcome and introduction to the event.
  • Walk through many themes in AI (hype, reliability, ethics, etc) via the use of headlines; includes a quick Mentimeter poll on 'AI optimism'.
  • Look at the historical development of AI, starting with old-but-familiar headlines about robots, via expert systems and the Dartmouth meeting, to deep learning and 'AI everywhere'.
  • Segment on ChatGPT in particular and Equinor's chat tool, mentioning information security considerations and also a compelling use case; there is a spot prize in this segment.
  • Survey of some real applications of 'domain AI' (mostly task-specific discriminative machine learning systems) vs 'workplace AI' (eg chat) in Equinor today.
  • Turn to the risk and ethics of AI, especially generative AI, and a look at the EU AI Act and what it might mean for us as citizens and as employees; segue back to Communications.
  • Quick wrap-up and an invitation to ask questions.

The presentation typically took 35 minutes or so, leaving about 15 minutes for a Q&A. There were always plenty of questions.

Outcomes

  • We reached more than 3000 employees with the roadshow, starting conversations about AI all over the company.
  • We were able to gather a large number of questions from the audience, which was the basis of producing a FAQ document.
  • The event helped us reach out to other partners in Digital Norway, and the broader community through LinkedIn posts and an article in Digi (see below).
  • The relationships and collaborations in the delivery team were valuable side-effects.

Reference