Despite the hoopla about AI displacing vast quantities of human workers, the reality to date is that the best results have been achieved not by man or machine working independently, but by them working together.
In Human + Machine, Accenture’s Paul Daugherty and James Wilson highlight three ways AI technologies can augment the work of humans in the workplace.
Amplification refers to the ability of AI to existing human capabilities even stronger by allowing us to tap into data-driven insights using real-time data.
For instance, Autodesk’s Dreamcatcher software uses AI to iterate through a range of possible designs to help the human designer produce something innovative. The grunt work is given to the machine, whilst the creative work remains with the human.
The second way they propose that AI tools will improve the way we work is via interaction. This is where AI technologies are used to facilitate interactions between people, or indeed on behalf of people.
A good example of such changes are Aida, the virtual help desk provided by Swedish bank SEB. The system was heavily trained so that it could confidently interact with one million customers. It now answers the easier questions, leaving human support staff to tackle the more challenging problems.
The final category takes AI from the virtual, software driven realm very much into the physical realm and deploys it in tangible, physical spaces. It’s the use of AI alongside sensors, motors and actuators that power the smart robots that work alongside us. Whereas in the past, such machines worked largely independently, the new generation are capable of working and collaborating with humans.
Such partnerships are especially common in the car industry, where state-of-the-art manufacturing lines feature lightweight, context-aware robot arms and cobots that are designed to work alongside humans.
These skills sit alongside three ways in which humans can improve and complement machines.
We’re moving on from an age in which humans adapt to how computers work, with the reverse increasingly the case. In order for this transition to occur however, we will need people to effectively train machines to work alongside us.
This could involve ensuring the data that machines are trained with is suitable, or correcting errors and reinforcing successes in machine behavior.
AI systems are taking on ever more important roles, and as they do so the ability to explain their working becomes ever more important. Being able therefore to shed light on the black box workings of an algorithm will be key.
Humans may therefore test, observe and explain the algorithms, or augment their interface to make them more explainable. It also seems likely that there will be a sizeable role for interpreting machine outputs, especially in areas such as healthcare.
There is also likely to be a significant role for people who will ensure that AI-systems operate as they should do. It’s crucial that AI serves us rather than the other way round, and people in this category will fulfil that role.
Tasks might include setting limits for AI-systems or flagging errors in machine judgement. It might even include a ‘machine resources’ department in the same that we have human resources departments now, who will be tasked with assessing and evaluating performance levels.
Whilst much of the narrative around the impact of AI on the workplace has revolved around the destructive capabilities of the technology, in reality there are still skills shortages throughout the economy. These shortages are only likely to increase as new skills are required in the ‘missing middle’, or the interface between man and machine.
“Our research has found that the real issue isn’t simply that humans will be replaced by machines; it’s that humans need to be better prepared to fill the growing number of jobs in the missing middle,” Daugherty and Wilson say.
Whereas most of the studies that have examined this issue to date have used very blunt instruments, the work outlined in Human + Machine is pleasingly grounded in what’s happening here and now. The authors don’t profess to suggest that no changes are required, but they are very much changes within our control. The challenge now is upon us to meet those challenges.