The continued spread of AI technology has stoked fears of catastrophic job losses in which machines and algorithms run the world. Employees worry they will soon report to robots—or be replaced by one.

But research conducted at Accenture paints a more encouraging picture in which people and machines are not rivals fighting for jobs. This more positive view of the future is based upon the powerful collaborations that take place when humans and machines work together, each taking advantage of the other’s complementary strengths.

Keep in mind that what’s easy for humans (folding a towel) can be devilishly tricky for machines. And what’s easy for machines (spotting hidden patterns in huge datasets) can be extremely difficult for us to do. As a result, many existing jobs are transforming while new ones are emerging in what we call the “missing middle,” where humans and machines work together.

Succeeding in this emerging space requires some adjusting. In our ongoing work and research at Accenture, we see evidence of at least eight novel skills that draw on the fusion of human and machine talent within a business process to create better outcomes than achieved by working independently.

While these skills are about profound new forms of human–machine interaction, they don’t necessarily require expertise in machine learning or URScript (the programming language for many industrial robots) or other technical areas. What’s needed instead are thoughtful humans who are eager to adapt these fundamental skills to the specific needs of their business.

1. Taking Back Your Time—and Doing Something Positive with It

With AI systems taking over many rote job tasks, workers must develop the ability to redirect their time to more distinctly human tasks like interpersonal interactions, creativity, and decision-making.

Consider the field of medicine. AI can help read X-rays and MRIs, find elevated risks of heart failure buried in medical records when a routine exam might miss it, and even help identify cancerous moles that might go undetected. All these insights can give physicians back precious minutes of human-to-human interaction with their patients. But that means physicians must be able to quickly shift gears toward patient care when AI gives them back their time.

2. Anticipating Unintended Consequences

The gap between the use of AI and the public’s broad acceptance and understanding of AI is massive. Bridging this gap is where the skill of normalizing the technology comes into play. Broad groups of people must become comfortable using AI.

One major event—such as a child being injured by a driverless car or truck drivers striking in opposition to the threat of self-driving trucks—could create a societal crisis of confidence in the technology at large. That’s why CEOs must anticipate resistance—through understanding the needs and concerns of communities affected by the changes wrought by AI—and find ways to ameliorate strife. Workers, too, can become allies in normalizing the technology if they feel executives and management are addressing their concerns and if they have some say in the matter.

This skill requires a subset of other skills, such as an understanding of the humanities, STEM skills, an entrepreneurial spirit, public relations acumen, and an awareness of social and community issues.

“Anyone who’s working in conjunction with an AI agent will have to act as a ‘role model’ to their digital colleagues.”

3. Using Your (Human) Judgment

AI can get many things right, but it still doesn’t know how to read situations and people well enough. Because of this, human judgment and effectuation will be a necessary new skill.

When Royal Dutch Shell sends its robots to monitor equipment and perform safety checks in the company’s remote facility in Kazakhstan, it still requires the know-how of workers who are on the lookout for hazards. The robot, called Sensabot, is the first of its kind approved for use by oil and gas companies in potentially dangerous environments. But a remote worker operates Sensabot, watching the video feed, making judgments about risks, and resolving ambiguities.

4. Asking the Right Questions

Knowing how best to ask questions of AI, across levels of abstraction, will be critical for human workers.

Steve Schnur, who runs retail operations at a major resort, uses AI from Revionics to optimize prices in the resort’s convenience stores. Even a small adjustment in the price of Advil or Band-Aids produces a significant effect—something that would have been impossible to understand (and ultimately control) without AI and an operator asking the AI system smart questions.

Schnur’s team uses the system to find the best prices of Advil, Band-Aids, soda, and more at any given time, under any number of constraints, based on weekly sales reports on about 7,000 items. Schnur poses the question: “If you raise the price of Advil, what happens to Tylenol?” The system can figure out the relationship between Advil and Tylenol, even though they’re only classified by stock-keeping units, and show that, for instance, the last time Advil was raised by 25 cents, sales of Tylenol increased. The system also lets Schnur probe pricing decisions in other ways, too, like “show me the most beneficial price changes” and “tell me which items will sell the least with a price increase.” The smarter the questions, the more insight he can gain and the better the picture AI can paint for him of his overall operations.

5. Helping Bots Help You

Workers must help bots help them extend their own human capabilities, business processes, and even careers. There are scheduling agents such as Clara and x.ai. There are tools to organize regular meetings so that you can mimic “chief of staff” activities with bots in Slack, like Howdy, Standup Bot, Tatsu, and Geekbot. You can share meeting minutes and highlight keywords with tools like Gridspace Sift and Pogo. You can improve writing with Textio or IBM’s Watson Tone Analyzer. And you can even have a bot post updates or pictures to social media to build your professional and personal brand on your behalf with Doli.io.

Bot-based empowerment skills also come in handy for searches. Job searches and recruitment are often a numbers game, so you’re already behind if you haven’t automated various elements of the process through services like LinkedIn or up-and-coming AI-based job search assistants like Wade & Wendy or Ella.

6. Understanding How Machines Work and Learn—and Adapting Accordingly

Ever used a tool that’s felt so familiar that it was as if the tool were an extension of your own body or mind? Melding occurs when you parallel park your car without assistance or swing a tennis racket to make contact with a ball. Increasingly, machines are getting better at melding with us as well. When you start typing search terms, Google not only considers the most generally popular associations for its autocomplete feature but also considers your geographic location and previous search terms. It can feel as if the software is reading your thoughts.

In the age of human–machine fusion, holistic (physical and mental) melding will become increasingly important. The full reimagination of business processes only becomes possible when humans create working mental models of how machines work and learn, and when machines capture user behaviour data to update their interactions. With melding, processes become flexible, adaptable, and potentially fun, like dancing with a skilled partner, switching from time to time the roles of lead and follow.

7. Being a Willing Learner—and Teacher

Apprenticing as a fusion skill marks a distinct break from the way we’ve historically managed technology. Traditionally, technological education has gone in one direction: people have learned how to use machines. But with AI, machines are learning from humans, and humans, in turn, are learning from machines. Apprenticing means customer service representatives or anyone who’s working in conjunction with an AI agent will have to act as a “role model” to their digital colleagues. This requires appropriate mentoring and technical skills, in addition to building AI to be easily trainable.

8. Reimagining… Just About Everything

The final—and perhaps most important—hybrid skill is the ability to reimagine how things currently are. Take Stitch Fix. The company is completely reimagining online sales and order fulfilment processes. Users of the service don’t actually shop for clothes; in fact, Stitch Fix doesn’t even have an online store. Instead, customers fill out style surveys, provide measurements, offer up Pinterest boards, and send in personal notes. Machine learning algorithms digest all of this eclectic and unstructured information. An interface communicates the algorithms’ results along with more nuanced data, such as personal notes, to the company’s fashion stylists, who then select five items from a variety of brands to send to the customer. Customers keep what they like and return anything that doesn’t suit them.

This approach illustrates three lessons about how to combine human expertise with AI systems. For years, the dream was to create an artificial intelligence that could rival that of people. Today we’re seeing that practical AI is becoming a tool to extend our own human capabilities at work. In turn, we’re guiding AI systems to evolve into better tools that further extend our capabilities. Never before in history have our tools been so responsive to us, and we to our tools.