Page 138 - Cyber Defense eMagazine January 2024
P. 138

Four Ways Genai Will Change the Contours


            of The Corporate Landscape In 2024


            By Neil Serebryany, CEO and Founder of CalypsoAI


            Generative artificial intelligence (GenAI) models, including large language models (LLMs) have been the
            focal point of the business world’s attention  since ChatGPT  made its debut just a year ago. They have
            revolutionized operational practices across sectors, from streamlining supply chains to enabling unique,
            detailed  customer  interactions.  While  not quite  ubiquitous  yet,  this technology  is getting  closer  to that
            milestone every day, and its potential for innovation is boundless. It’s clear these models and their other
            GenAI  cousins  are  poised  to  reshape  the  corporate  landscape  even  further.  Here  are  some  ways  I
            anticipate they will do so in the upcoming year.

               •  The first  large-scale  breach  of a foundation  model provider,  such  as OpenAI,  Microsoft,
                   Google,  etc.,  will  happen  in  the  upcoming  year  and  will  lead  to  a  large-scale  security
                   incident.  The scope and  scale of the attack itself  will be on par with  recent incidents,  such as
                   Microsoft’s  “accidental”  disclosure  of  38  terabytes  of  private  data  and  Google  Fi’s  hack  that
                   exposed the data of 38 million customers. With the amount of sensitive information that has been
                   sent to LLMs like ChatGPT, the fallout would be profound and could easily exceed either of those




            Cyber Defense eMagazine – January 2024 Edition                                                                                                                                                                                                          138
            Copyright © 2024, Cyber Defense Magazine. All rights reserved worldwide.
   133   134   135   136   137   138   139   140   141   142   143