It’s been one year since the public release of ChatGPT and since then a lot of discussions have taken place about the impact of the technology in the workplace. Some of the discussions were triggered by research. In this post, I offer 5 insights from the research published this year and conclude with 5 implications for corporations.
Generative Artificial Intelligence (AI) has emerged as a transformative force in the world of technology and business. At its core, generative AI refers to a class of algorithms and models designed to generate human-like text, images, and even sounds autonomously. Over the past year, this field has witnessed remarkable advancements, pushing the boundaries of what AI can accomplish.
With models like GPT-4, generative AI has become more proficient at understanding and generating human-like content. These models are capable of tasks such as natural language understanding, text generation, language translation, and more. These advancements have opened up a plethora of possibilities for applications in various industries, from content creation and customer service chatbots to personalized marketing and healthcare diagnostics.
Beyond the technical progress, the level of adoption is particularly striking. It took two months for ChatGPT to reach 100 million users and at the end of 2023, the service scored 100 million weekly users. In this blog post, I will delve into the latest research and insights regarding the impact of generative AI on the workforce.
5 insights from research
- Enhancing productivity with generative AI. Recent studies underscore the significant impact of generative AI on workforce productivity. Noy and Zhang’s research involving 444 college-educated professionals revealed that exposure to ChatGPT led to a notable increase in productivity. Participants who used ChatGPT showed a reduction in time taken by 0.8 standard deviations (SDs) and an improvement in output quality by 0.4 SDs. Similarly, a study with 758 BCG consultants demonstrated that those with access to GPT-4 AI completed tasks 12.2% more and 25.1% faster on average than those without AI tools.
- Quality enhancement alongside productivity. Contrary to concerns that increased productivity might compromise quality, both studies reported positive effects on output quality. For instance, AI-assisted consultants in the BCG study produced results that were over 40% higher in quality compared to a control group. Additional research also highlights the proficiency of LLMs in providing constructive feedback and ideation, often rivalling human capabilities.
- Amplified benefits for lower performers. An interesting observation from these studies is the disproportionate benefit of generative AI for lower-performing individuals. In both the Noy and Zhang and BCG studies, those below-average performance benchmarks saw greater improvements in productivity and quality, suggesting that AI tools might help in levelling the playing field.
- The dual role of generative AI: Automation and job transformation. Noy and Zhang’s research suggests that generative AI primarily substitutes for worker effort in certain tasks, shifting the focus towards more creative aspects like ideation and editing. However, tasks outside AI’s current capabilities showed a decrease in performance when AI tools were employed. A separate study on the impact on freelancers in fields like copywriting and graphic design indicates a short-term reduction in demand for these professions, highlighting the disruptive potential of AI.
- 5- Diverse perspectives on AI adoption. Studies on AI adoption present a complex picture. One study found that 33-46% of crowd workers utilized LLMs like ChatGPT for task completion. Salesforce’s survey across 14 countries revealed that over half of the workers use generative AI without official permission, with 40% using prohibited AI tools. Furthermore, a study on AI in medical diagnosis showed that collaboration between AI and human experts did not yield better results, possibly due to the underutilization of AI insights by professionals.
5 Strategic implications
- Optimizing task allocation in the era of generative AI. A critical aspect of integrating generative AI into the workforce is discerning which tasks are best suited for automation. The study by Noy and Zhang illustrates a significant shift in job roles, with a notable decrease in the time spent on initial drafting and an increase in the time devoted to editing. Identifying the right balance between AI and human input becomes crucial for maximizing efficiency and effectiveness. It’s paramount to identify the tasks which could and (should not) be supported by generative AI.
- Generative AI as a catalyst for competitive reshuffling. The adoption of generative AI can radically alter the competitive landscape. Organizations adept at redefining their processes to integrate generative AI stand to gain a substantial competitive edge. This technology not only poses a threat to established market leaders but also offers them an opportunity to consolidate their position by embracing AI-driven transformations.
- The growing importance of data quality. Recent findings suggest that using AI-generated content in model training can lead to significant biases in the models, emphasizing the importance of data quality. The ability to ensure that humans perform critical tasks unaffected by AI biases is emerging as a competitive differentiator. The challenge lies in effectively implementing strategies that uphold data integrity and quality.
- Emphasizing change management for enhanced AI adoption. The synergy between technical and human assets is key to realizing the full potential of AI. Success in this domain not only depends on redefining processes but also hinges on fostering an organizational culture that understands and trusts AI. Open dialogues about the technology’s capabilities, limitations, and optimal use cases are essential to develop a shared vision for AI integration.
- The imperative of AI-focused training programs. Training in generative AI technologies is becoming increasingly vital, especially for enhancing the skills of lower performers. As AI continues to commoditize certain aspects of knowledge work, it’s not just the low performers who are affected; skilled professionals are also at risk. Comprehensive training programs can help employees understand the nuances of AI, including its limitations and best practices. This approach aligns with the insights from our previous blog on ‘When AI Decreases Performance’, emphasizing the need for strategic management of AI in corporate settings.
Generative AI is not just a technological advancement; it’s a paradigm shift in how we work and compete
Implications for the corporate world are profound and multifaceted. From dramatically enhancing productivity and quality to reshaping competitive landscapes and job roles, generative AI is not just a technological advancement; it’s a paradigm shift in how we work and compete.
The key lies in judiciously identifying tasks for AI integration, prioritizing data integrity, and embracing change management to foster AI adoption. Training and continuous learning emerge as crucial strategies to equip the workforce for this new era, ensuring that both high and low performers adapt and thrive. Understanding these dynamics and preparing for them is not just an exercise; it’s imperative for future success.
This article is based on a post published on Prof. Benyayer’s blog.
This post gives the views of its author, not the position of ESCP Business School.