Back to Liminal Blog

Generative AI Dramatically Increases Productivity

Explore data around how generative AI tools enhance productivity, and what those enhancements mean to organizations.

Share On:
Share on LinkedIn
Share on Twitter

Generative AI is transforming how businesses and industries operate. From automating labor-intensive tasks, to expediting complex processes, to fostering creativity and innovation, generative AI represents the greatest productivity leap in a generation. This article explores data around how these tools enhance productivity, and what those enhancements mean to organizations. 

Measuring Productivity Impacts of Generative AI 

Two key factors come into consideration when examining productivity gains driven by generative AI. The first concerns increased efficiency directly linked to the ability to accomplish more tasks at a faster pace with generative AI-enabled tools. Secondly, there is a strategic aspect involving resource allocation, and empowering individuals to focus their time and effort on tasks where human capabilities excel. Virtually every role in every industry includes functional responsibilities that can be automated or expedited with generative AI. The assessment of productivity also considers the increase in capacity achieved by automating rote and repetitive tasks.

More. Faster. Better. 

Three recent research efforts conducted across industries, roles, and functions highlight the impact of generative AI on task completion.

The first analysis, conducted by Harvard Business School, pitted two cohorts of management consultants against each other: a test group that was provided access to GPT-4, and a control group that was not. Outputs were then measured based on each group’s performance in completing a set of specific tasks, including rapid idea generation, persuasive writing, strategic analysis, and product innovation.The results show that the consultants with access to GPT-4 completed 12.2% more tasks, completed them 25.1% more quickly, and did so at a perceived 40% higher quality than the control group. 

In the second assessment, a group of researchers at MIT and Stanford University conducted a joint study on task completion, this time for customer support representatives at a large software firm. Researchers found that support agents with access to generative AI tools were able to handle 13.8% more inquiries per hour than agents without AI assistance. Within those results, the study further found that the least skilled representatives benefited the most, with the lowest quintile of agents increasing task completion speed by 35%.

In the third study, conducted by a team from Microsoft Research, GitHub, and MIT, researchers  explored the impact of generative AI on task completion for developers. In this test, a group of engineers was tasked with implementing an HTTP server in JavaScript. Half of the developers were provided access to GitHub’s generative AI Copilot, and the other half did not use any generative AI tools. Their findings found that the group with access to generative AI realized a 126% increase in task throughput.

In each instance, workers equipped with generative AI tools were able to drastically outperform those without access. 

Leveraging Human Capital More Efficiently

In addition to greater task outputs, generative AI can drive increased productivity by enabling  process reorganization to ensure resources are focused most appropriately. According to McKinsey, workers are spending as much as 60% to 70% of their time on activities that could be automated with generative AI. By eliminating repetitive or tedious functions, skilled employees can dedicate more of their capacity to those critical tasks that demand human engagement. 

Data Security is the First Step in Any Generative AI Journey

Generative AI is a productivity game changer, but it comes with real risks associated with data security, privacy, and sovereignty. Firms must take steps to prevent the inappropriate internal sharing or inadvertent external leakage of sensitive company, customer, and employee data in interactions with generative AI. By implementing a multi-layered security approach that integrates policy, process, and technology, cybersecurity teams can help their organizations say yes to the most promising productivity-driving projects on their radar.

Liminal is the technology security layer for companies looking to leverage generative AI. Click here to schedule a demo of the Liminal Platform.