Preparing for Artificial Intelligence
“Artificial intelligence (AI) systems are not just automating many processes, making them more efficient; they are now enabling people and machines to work collaboratively in novel ways. In doing so, they are changing the very nature of work, requiring us to manage our operations and employees in dramatically different ways.” So write Paul R. Daugherty and H. James Wilson in their book Human + Machine: Reimagining Work in the Age of AI (Boston: Harvard Business Review Press, 2018).
Daugherty is chief technology & innovation officer at Accenture (accenture.com); James Wilson, the managing director of information technology and business research at Accenture Research. Their book focuses on the “missing middle”—the traditional binary approach of pitting humans against machines and “fighting for the other’s jobs,” replacing that “humans and machines work[ing] together as allies to take advantage of each other’s complementary strengths.” This approach requires “think[ing] of AI as an investment in human talent first and technology second.” It includes “reimagining business processes” that can benefit from the collaboration of people and machines (hence the title), recognizing and creating AI-based jobs, and reskilling the workforce.
Waves of automation
The first wave of business transformation, explain the authors, “involved standardized processes.” Think Henry Ford. “The second wave consisted of automated processes.” Think information technology starting in the 1970s and business process reengineering through the 1990s. “Now, the third wave involves adaptive processes, which builds on the previous two. This adaptive capability is being driven by real-time data rather than by a prior sequence of steps. The paradox is that although these processes are not standardized or routine, they can repeatedly deliver better outcomes.”
AI systems gradually replacing humans in one industry after another across the board is, claim Daugherty and Wilson, “a widespread misconception.” AI-based systems do “what they do best: performing repetitive tasks, analyzing huge data sets and handling routine cases. They are amplifying our skills and collaborating with us to achieve productivity gains that have previously not been possible.” With AI, self-aware robot and robot arms—“kinder, gentler robots”—can now co-exist—work together—and collaborate with humans.
The “great irony” here, the authors continue, “is that some of the most-automated environments—the factory and other industrial settings—are experiencing a renaissance of human labor. In many cases, AI is freeing up time, creativity and human capital, essentially letting people work more like humans and less like robots.”
When robots are able to coordinate among themselves, says Gabriel Seiberth, managing director in automotive at the Munich office of Accenture Digital, in a separate conversation stemming from the book, “they don’t need to follow a strict proscribed plan; they can be flexible on specific tasks.” As a result, intelligent systems will “increase [manufacturing] capability, and by that, the level of customization. Production side customization, if you will.”
New jobs coming
Daugherty and Wilson predict AI will create eight new “fusion skills” that will lead to “entire categories of different jobs.” Trainers will help natural-language processors and language translators make fewer errors; train AI algorithms to mimic human behaviors (such as empathy and global perspectives); and help intelligent machines go beyond fundamentals (“further nuance and resilience”). Data hygienists will cleanse data beyond the data consistency and completeness required for data-driven operations. “Not only do the algorithms themselves need to be unbiased, but the data used to train them must also be free from any slanted perspective.”
Explainers will “bridge the gap between technologists and business leaders” and “will become more important as AI systems become increasingly opaque” in their “black-box nature of sophisticated machine-learning algorithms.” Algorithm forensics analysts can “autopsy” AI actions to uncover the impetus for unintended consequences. Similarly, transparency analysts can “classify the reasons a particular AI algorithm acts as a black box.” (For example, some algorithms purposely protect intellectual property.)
Sustainers are a third job category. Context designers balance the “variety of contextual factors, including the business environment, the process task, the individual users, cultural issues and so forth” when developing new AI systems. AI safety engineers “try to anticipate the unintended consequences of an AI system and also address any harmful occurrences with the appropriate urgency.” Last, “ethics compliance managers act as watchdogs and ombudsmen for upholding generally accepted norms of human values and morals.”
Ethical AI should not be taken lightly. While companies are profit driven, and “of course they will do everything they can to maximize their profit,” says Seiberth, he believes the overall AI environment “will be more and more ethics driven. This will impact the overall making of organizations. You can see that happening now.” (Example: Facebook. Another: Google, in participating in the Department of Defense Maven program, which weaponizes AI.)
Consequently, write Daugherty and Wilson, there will be automation ethicists and machine relations managers. The latter act as human resources managers, “except they will oversee AI systems instead of human workers. They will be responsible for regularly conducting performance reviews of a company’s AI systems. They will promote those systems that perform well, replicating variants and deploying them to other parts of the organization. Those systems with poor performance will be demoted and possibly decommissioned.”
AI really is different
Implementing AI is not the same as implementing other information technologies. True, says Seiberth, it’s “the same in terms of the requirements. You need data, you need to cleanse data, you need to quickly convert data.” But the information technology “has to be able to learn and adapt, and find solutions to problems we can’t explain in a precise way. This is completely different than what we’ve seen before.”
This is in fact what AI is really promising, Seibert continues. “You can automate tasks that you can’t operationalize at the time. For example, driving a bike. There is something you can’t explain. You know what it looks like, and you know how you can feel it, but you can never analyze or formalize it in a way that another person can learn [to bicycle] just from reading. This will present completely new challenges and tasks and roles.”
“The problem right now,” conclude Daugherty and Wilson, “isn’t so much that robots are replacing jobs; it’s that workers aren’t prepared with the right skills necessary for jobs that are evolving fast due to new technologies, such as AI. And the challenge will only grow as companies apply AI and reimagine work in other areas.”
When you think of developing maps for automated vehicles, you might think along the lines of something that you see on your phone, on a screen centered in your IP or something from Garmin.
Machine learning through pattern recognition can be a faster, less-expensive approach to monitoring manufacturing.
Mountain View, Calif.–based startup Drive.ai is launching an autonomous vehicle pilot program on public roads in Frisco, Tex.