“ With foundation models , the proper governance to enable responsible AI is vital ”
ALESSANDRO CURIONI IBM
CREDIT : IBM
At the end of 2023 , the corporation released watsonx . governance to help businesses automate responsible , transparent machine learning and generative AI ( Gen AI ) workflows . IBM says that the platform will help businesses eliminate the mystery around the data going in and the answers coming out of their AI models .
Whilst AI models at this scale have a wide variety of use cases , Curioni stresses the importance of governance to IBM . “ With foundation models , the proper governance to enable responsible AI is vital ,” he says . “ Once you have a big model , everything bad is transferred to other small models that you create from that .”
Developing and harnessing an AI model that is free of bias is the end goal , as businesses seek to be more transparent to identify and tackle these issues when they arise .
Curioni says : “ That ’ s why it ’ s important to have a technology that allows not only governance but dynamic governance . Now , you have the possibility to continuously test how your models behave across several dimensions . You need to have a way to bring out a model that is compliant – that is our approach .”
IBM ’ s latest milestones in foundation models Curioni highlights that IBM ’ s priorities are digital labour , which means creating technology that is able to automate business
aimagazine . com 57