Shutterstock
While AI remains in a “hype cycle,” one of the presenters at the Texas Association of State Systems for Computing and Communications (TASSCC) annual conference reminded attendees that AI isn’t new but has been in various cycles for at least three decades.
Knowledge creation has been moving from human intelligence to synthetic systems that include statistical learning and now contextual adaptation.
Surrounding issues include privacy, copyright, data security, transparency, accountability and accuracy, said Gita Lal of Daman Consulting, who introduced the final session featuring Microsoft and Google representatives working in data and AI.
Among the takeaways:
- Document summary is where many users of AI are starting. It is a smaller step that is controllable within a dedicated environment.
- A dedicated environment, such as an enterprise cloud that is augmented with AI, allows IT professionals to use their own vetted data or documents without releasing information into the open web.
- Using open web tools to experiment with AI tools means that information becomes part of the larger AI learning environment, so users must be mindful of how data is used and stored for privacy purposes.
- Data governance goes hand in hand with AI governance because AI draws from massive data sets that need to be clean and curated, as in “garbage in, garbage out.”
Lal said she believes that IT leaders will need to develop staff training on AI and AI models so everyone better understands the landscape and be able to interact with AI because it will always require human interaction.
