Technical Architects, Stay Ahead of the Top 6 GenAI Security Risks
This Event was presented by Dennis Xu the Sr Director Analyst at Gartner and Hemant Narang Sr. Gartner Specialist. The spoke on the topic of AI, the pros and cons, and also its limitations.
A few points that was spoke about in the presentation that stood out to me was the growing danger of data insecurity as open source AI is being fed more and more data. For example ChatGPT 3.5 with the trick of a few loop questions would give out private information, this error is currently not well understood according to the presenters and its being improved upon. (This is presented on the left)
A positive about AI that was spoke about in the presentation is its accessibility to anyone who uses the internet not just on browser but also application, There is more to be done on accessibility, and this is evolving everyday with more and more user of ChatGPT. There was also talk about the Toxic effect of data hallucination in which ChatGPT is not intuitively smart enough to give an answer to a question that it doesn’t have an answer for, or if there is multiple sources for the same answer.
Although this was a short presentation I believe this is beneficial to those going into the IT field in understanding how to utilize AI practices. Being aware of security and information malfunction is also crucial to use any type of tool, this presentation is beneficial to my personal professional journey because of my need to evolve with the knowledge of the world and the new things it brings.