The Task-Based Approach to Preparing for the Future of Work

This month, Singapore continued its sustained progress in AI adaptation with the launch of A Guide to Job Redesign in the Age of AI. Amidst the COVID-19 pandemic, digitisation and timeframes for the adoption of AI have accelerated.

A variety of Singaporean institutions contributed to the Guide, including: the Advisory Council of the Ethical Use of AI and Data, the Infocomm Media Development Authority (IMDA), the Personal Data Protection Commission (PDPC), and the Lee Kuan Yew Centre for Innovative Cities at the Singapore University of Technology and Design. The Guide outlines what measures organisations might implement to incorporate AI while building trust among employees. Recognising a need to be prepared for the future of work, the Guide advocates a task-based and human-centric approach to the adoption of AI.

“AI overlaps with and expands upon automation, allowing for transformations to occur along a continuum of specific tasks, stretching from the repetitive to the cognitive.”

The task-based approach is an effective tool for organisations to assess the impact of AI on work and employees. By making assessments on the interaction and impacts of tasks, organisations can tailor conscious plans to prepare for the future of work. In this regard, the Guide identifies four areas where organisations could adopt measures to redesign jobs.

1. Transforming Jobs

The first section of the Guide seeks to assist organisations in assessing how AI will affect work with task-based job transformation roadmaps. In making such assessments, organisations can pre-empt how AI might augment, support, or replace employees responsible for specific tasks. This reflects a six-stage process:

  1. Break down jobs into tasks: breaking a job into constituent tasks allows organisations to build a detailed and practical risk profile for the role that shows which tasks can be replaced or augmented by AI.
  2. Assess the potential impact of AI on each of the tasks: organisations should work with internal and external experts to assess how well (or when) an AI solution will be able to perform a task.
  3. Assess if AI should be implemented for each task, and the extent to which AI can be deployed: to promote trust and understanding in the use of AI, organisations should ensure the decision-making process is explainable, transparent, and fair; and that AI solutions are human-centric. Additional consideration should be given to the critical thinking, creativity, and emotional intelligence requirement for tasks.
  4. Consult managers and employees about which tasks are valuable to them: by asking employees about the tasks they value, organisations can attain insight as to how well a proposed AI solution fits the task they perform.
  5. Decide the appropriate time frame to implement AI: consideration should be given to a) how much change can be implemented at once; b) capability to support the use of AI; c) resources to support the transformation; d) organisational culture to change; and e) employees’ readiness for AI.
  6. Recombine and reconstruct the transformed tasks into a future job: a future may combine tasks that could be automated; tasks for which AI should augment the employee; and/or tasks that should remain in human hands.

2. Charting clear pathways between jobs

Once organisations have determined how AI might affect work, they may begin preparing task pathways between jobs and identifying the tasks employees would need to learn to transition to other jobs. This reflects a distinction between tasks and skills. While tasks fit within the skills framework, the generality of skills leaves much to be desired when preparing for the future of work. When charting task pathways for employees, organisations should:

  1. Standardise definitions of tasks across jobs and competency or skill levels;
  2. Use shared similar tasks to generate potential pathways to other jobs;
  3. Identify the tasks that are not similar for training and skills development; and
  4. Re-examine assumptions about jobs to begin changing mind sets.

3. Clearing barriers to transformation

The Guide identifies two primary barriers to the adoption of AI in the workplace: technical and human. These barriers may be multi-dimensional with implications in organisation, occupation, technology, and phase of transformation. Organisations must aim to identify the location and extent of these barriers to provide effective remedies. Acknowledging these potential barriers enables organisations to establish appropriate plans such as:

  1. Creating a dedicated team of innovation champions to identify pain points and opportunities;
  2. Phasing out implementation and providing training in small batches to ensure business is as usual;
  3. Providing relevant training and learning for the precise tasks(s); and
  4. Ensuring that the training would be clearly applicable and available in an accessible form.

4. Enabling effective communication between employers and employees

Finally, organisations must aim to build trust for AI among employees. Historically, if employees’ perceive themselves to be at risk, they will show greater resistance to change or new technologies. Organisations must proactively communicate with their employees to work to identify and prevent rifts appearing. Importantly, there must be communication along three lines: “why” AI is beneficial, “what” is the process for implementation, and “how” can employers address the concerns of their employees.

With the impact of the COVID-19 pandemic, the timeline for the incorporation of AI has been accelerated. Rapid digitisation is pushing organisations and employers to accelerate their timelines, potentially to the detriment of preparedness and employee trust. Singapore’s Guide to Job Redesign highlights that by taking a task-based approach, organisation’s may improve the incorporation of AI while simultaneously building trust in its capabilities.

You can find the Guide here.

Posted in AI