Key Points

1. AI augmentation is the collaborative use of AI systems to enhance, support, and amplify the cognitive and physical capabilities of human workers, rather than replacing them entirely. The purpose is to increase productivity and quality of output by enabling humans to work faster and smarter

   
   

2. AI augmentation is a safe way to carefully and gradually include AI as a collaborator

   
   

3. There are several steps to achieving AI augmentation, starting with identifying the repetitive tasks that can be automatable 

AI augmentation is the collaborative use of AI systems to enhance, support, and amplify the cognitive and physical capabilities of human workers, rather than replacing them entirely. The purpose is to increase productivity and quality of output by enabling humans to work faster and smarter. Compared to full automation, augmentation is about giving existing valuable staff superpowers.

You may have heard of collaborative robots, also known as cobots, which are industrial robots that can safely operate alongside humans in a shared workspace (unlike autonomous robots, which are hard-coded to repeatedly perform one task, work independently and remain stationary). In short, the goal is to combine the strengths of the AI with those of the human.

What is an Example of AI Augmentation?

For example, if someone needs to draft a proposal, that person could combine their abilities with AI’s capabilities. That is, the writer can decide which reports to select to include for the coverage in the proposal, and then ask the AI to list five of the most impactful statistics from those reports. At this point, the writer could ask the AI to draft a first draft of the proposal with those five set of statistics. From there, the writer could edit the document and complete an ethics check at the end.

Together, the AI and the human writer could synthesize data, draft a document, edit the document, and do the final ethics check.

How do HR Leaders Achieve AI Augmentation?

AI Augmentation is the most responsible way to introduce AI. The reason is because it is not full automation, which can carry high risk and complexity, but it does not involve compiling statistics manually from multiple reports, which is the traditional way of doing things on the other end of the spectrum. AI augmentation is a happy medium.

In fact, this is a safe way to carefully and gradually include AI as a collaborator. The AI can do the things that it is good at like sifting through mountains of data, finding patterns, and completing the repetitive tasks that bores most humans. This frees humans to focus on what they can do best, such as using expertise to solve tricky problems, building relationships with customers, and thinking creatively and empathetically. An humans can perform final ethics checks too.

The following steps can lead to full AI augmentation, so that humans can still be in the driver’s seat instead of watching from the sidelines:

• Level 1: Use AI augmentation to eliminate the boring stuff. Identify the routine, automatable tasks in a job that slows everything down. Have AI start by taking on those tasks. For instance, the AI can clean up customer service tickets and thin out the queue

  
  

• Level 2: Allow workers to have AI tools that act as portable experts. Allow workers to use these experts to enhance the worker’s work quality and productivity. For example, the human customer service agent can ask the AI to read a ticket and respond by creating a first draft of a customer response. The human agent can review it, edit it, and confirm that it is an appropriate message before sending

  
  

• Level 3: Use AI augmentation for predictable tasks. Identify the more predictable tasks. Allow a more autonomous AI system to deal with specific predictable tasks, Predictable tasks could include things like answering the common question, “Where is my order?”, so that AI systems handle these type of tasks completely on their own—but if at some point where the AI system flags a more complex issue, the task escalates and the human agent can seamlessly take over the task—the human is always in control

What are Some Best Practices for Using AI Augmentation?

Here are a few tips that can help a businesses with AI augmentation:

• Use the knowledge and experience you have to train the AI system

  
  

• Remember to test the AI in a risk-free environment (a safe and stable sandbox)

  
  

• Make sure to roll out the AI slowly and make necessary adjustments

  
  

• Noted relevant metrics, measure the value created with AI augmentation, and note the value created by the AI-human collaborations

  
  

• Create training opportunities for employees with respect to AI-human collaborations

Conclusion

According to Gartner, it is necessary for HR leaders to plan for a blended workforce. This involves moving from a mindset where AI is viewed as a nice-to-have bolt-on to a regular practice of designing a human-AI workforce where both use their strengths and co-deliver work.

Moreover, EY recommends blending operational gains with a people-first mindset. More specifically, the chances of sustainable business and capability growth hinge on whether organizations keep a people-first mindset while integrating new technologies. To accomplish this, EY suggests that organizations deploy the most efficient tools and processes to create sustainable value while still investing in the skills, career and personal growth of the workforce to create a more exceptional employee experience.

This means bringing a holistic, people-centered perspective to an increasingly more digital world of work. While there may be a percentage of tasks for every employee that might be supported by AI tools, organizations will need those employees to be the human-in-the-loop who makes final decisions.

Finally, employers are recommended to:

• Appreciate AI’s role in a comprehensive workforce strategy, and be aware of the potential challenges that lie ahead

  
  

• Determine how AI can empower workers in the organization

  
  

• Explore potential risks and security concerns

  
  

• Consider size, scope and cost in terms of evaluating performance and cost trade-offs

  
  

• With regards to training on the new tools, chart the path forward with people at the center

  
  

• Implement metrics that measure workforce sentiment tied to confidence in and adoption of the new technology