Smart Machines Still Need Smart People
Smart machines are now capable of replicating many human capabilities. In a Deloitte Twitter chat, experts weighed in on the enterprise implications.
Amplified intelligence, the use of technology to augment human intelligence in the enterprise, has the potential to transform many traditional jobs and business capabilities. Machines are increasingly capable of completing tasks previously undertaken by humans. But amplified intelligence is still in its early days, and many CIOs and other executives are questioning how to prepare for the rise of the machines from both a talent and technology perspective.
Deloitte Analytics hosted a chat on Twitter and invited four luminaries to share their observations about amplified intelligence. As compared to artificial intelligence (AI), which refers to the capabilities of the machines themselves, amplified intelligence makes the most of both smart machines and smart people, the participants agreed, and signals the augmentation, not automation, of many traditional human workplace functions.
But while the enterprise possibilities are exciting, change can be unnerving. According to Bill Briggs, chief technology officer at Deloitte Consulting LLP, amplified intelligence “is fueling both opportunities for renewal and innovation, and existential threat of disruption.”
Changing Roles and Relationships
New breeds of smart machines can replicate many traditional human capabilities, including gathering and analyzing data automatically, and subsequently providing recommendations for next courses of action. But humans are still far better at evaluating actions and responses in context, challenging assumptions, and overriding machine-generated suggestions when circumstances demand it. “The best of both will rule the day,” says Forrest Danson, who leads the U.S. Deloitte Analytics Integrated Market Offering. By freeing up people to focus on decision-making rather than number-crunching, “artificial intelligence can enhance the effectiveness of knowledge workers, not just replace them.”
Anthony Abbatiello, a principal in Deloitte Consulting LLP’s Human Capital practice, agrees that machines are not likely to make skilled employees expendable anytime soon. “The need remains for leaders to translate data into meaningful development activities,” he says.
Tom Davenport, a renowned thought leader and author, takes a slightly different tack, noting that knowledge workers and other professionals do have reason to be concerned about their job security. “Their jobs won’t all go away, but we won’t need as many employees,” he says. To counteract potential job losses, he believes that humans must both work with smart machines—by building them, maintaining them, and assessing their performance—and work around them, by focusing their efforts on tasks that machines can’t perform, often in niche areas too narrow to automate. Employees who master both sides of this equation “will be working with smart machines as colleagues, not as bosses, not as slaves,” he says.
To prepare for this new reality, Davenport advocates readying employees for job changes to come, and understanding the advanced technologies available to them. “Some systems will be better with numbers, some better with text, some with learning, and some with static inputs,” he says.
Security and Privacy Implications
Some of the most hyped concerns about artificial intelligence relate to HAL—two of the chat experts referenced the sentient computer and primary antagonist in the science fiction novel and film 2001: A Space Odyssey as a cautionary tale. But rather than focus on machines as potential overlords, leaders should consider that AI amplified by human interpretation can help bolster enterprise security and privacy strategy, the experts agreed.
“People naturally gravitate toward the potential threats posed by AI and robotics, which are very real, but amplified intelligence can have a huge positive impact in the area of cybersecurity—driving resilience and vigilance, performing security and event monitoring to recognize normal, baseline behavior and activity, and proactively responding to anomalies,” Briggs says.
Danson says that protecting access to and control of smart machines will be critical; many technology vendors are already offering those products and services. At the same time, he believes that “using AI to identify real threats and to self-preserve will be part of the answer.” Davenport adds that many advanced machines will know when outsiders are tampering with them, and be capable of tipping off humans to heretofore unknown security gaps.
The Promise of Amplified Intelligence
So how are enterprises putting amplified intelligence into action? The experts came to the chat armed with a bevy of real-world examples of machine work bolstered by human intelligence.
Davenport points to the rise of robotic advisors in financial services, which operate according to manually entered algorithms, and IBM’s Watson robot, which is aiding humans in their efforts to cure and treat cancer. Indeed, AI is already spreading across health care, helping workers to diagnose and prescribe drugs more rapidly, Abbatiello says. And Danson notes that machines can identify billing anomalies and generate journalistic reports of events and activities, subject to human review and approval. While health care, financial services, and manufacturing are the first fields to seize these capabilities, others are on their way, he says.
Where might amplified intelligence take enterprises from here? “It will prompt us to get smarter.” Davenport says. “We’ll get rid of more boring work. I proffer that we are entering a brave new world in which we are not entirely sure what’s going to happen.”
Perhaps most importantly, amplified intelligence can encourage innovation by transforming how many companies complete work and reach business conclusions. It can help organizations overcome institutional inertia and trump the “just because” responses that can stifle new ideas, Briggs says.
“Amplified intelligence may finally put a fork in default decisions based on ‘We’ve always done it this way,’” Danson adds. “The next several years will be a wild ride.”
August 11, 2015, 12:01am
Questions? Write to Deloitte CIO Journal Editor
This publication contains general information only and Deloitte LLP and its subsidiaries (“Deloitte”) are not, by means of this publication, rendering accounting, business, financial, investment, legal, tax or other professional advice or services. This publication is not a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business. Before making any decision or taking any action that may affect your business, you should consult a qualified professional advisor. Deloitte shall not be responsible for any loss sustained by any person who relies on this publication. Copyright © 2015 Deloitte Development LLC.
Source: Smart Machines Still Need Smart People
Via: Google Alerts for AI