Microsoft Vice President highlights potential of generative AI tools to reduce costs, boost efficiency
(MENAFN) Microsoft's Vice President, Vik Singh, believes that generative AI tools have the potential to significantly save companies both time and money, despite the technology still needing improvements, particularly in recognizing and admitting its limitations. Singh pointed out that current AI models, such as those powering popular applications like ChatGPT, often fail to indicate uncertainty when they lack confidence in their responses. This issue can be problematic for businesses relying on AI-generated information, as these tools sometimes produce inaccurate or fabricated answers, known as "hallucinations." He emphasized that addressing this flaw is crucial since companies using Microsoft’s AI-based assistant, Copilot, would not tolerate even occasional inaccuracies.
Singh, who oversees the development of Copilot applications at Microsoft, explained that many experts are working diligently to help AI models better recognize questions they cannot answer accurately and prompt for human assistance when needed. He argued that an AI model with more modest capabilities could still offer substantial value. For example, he cited a case where an AI-based customer service assistant was used by a company that previously spent USD8 per query whenever a human agent was involved. Even if the AI could only handle half of the queries autonomously, the company still realized substantial savings.
Since joining Microsoft in January, Singh has led the development teams working on AI-powered tools for sales, accounting, and services. These specialized applications, although less mainstream, demonstrate the commercial potential of generative AI and justify the significant investments tech companies are making in this field. Singh noted that Microsoft's Copilot tool, for instance, helps conduct market research, freeing up time for sales teams to focus on customer engagement. One of the companies using Copilot, Lumen Communications, reportedly saves around USD50 million annually by leveraging these AI-driven efficiencies.
MENAFN09092024000045015682ID1108651180
Legal Disclaimer:
MENAFN provides the information “as is” without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the provider above.