Cutting through the AI hype: Explainable AI and a flexible workforce

(Image credit: Shutterstock/Mopic)

AI will undoubtedly change the way we work and the way we do business – but let’s keep the hyperbole to a minimum. An informed c-suite and educated workforce will help the transition to collaborative AI working practices. Understanding that AI applications will lighten the load for employees - who are weighed down by administrative tasks and require life-long learning in order to pivot between jobs as business accelerates - will be essential. But the true implications will be defined by the flexibility of businesses, their leaders and employees. Ultimately trust will be won by compliant, transparent and explainable AI practices.

Digital transformation is shifting the goal posts when it comes to the difference between simply surviving and positively thriving in the contemporary enterprise sphere. Enabling this shift towards data-driven business strategies will facilitate the addition of AI to the workforce, truly turbo-charging productivity. This is already happening across industries, but not as portrayed in the dramatic realms of sci-fi classics, or indeed popular tabloid newspapers. In more subtle and tangible ways, AI will create an opportunity for those impacted to gain and develop new and more diverse skills, and even give precious time back to us humans.

So, what is the AI reality, that sits comfortably behind the hype? Well, it is here already, improving working processes, and it needs to be taken seriously in order for businesses to remain competitive. However, it has not yet reached its full potential and offers huge opportunities for businesses of all sizes. According to a report from PwC, AI could contributes up to $15.7 trillion to the global economy by 2030. The race is on and China has made its intentions clear after announcing its desire to be a world leader in AI by 2030. Apart from creating over 10,000 AI research papers in 2015 alone, with the largest population in the world China has access to more data than any other nation – a prerequisite for training machine learning.

We can't be replaced that easily

However, the USA is leading the way with tech goliaths such as Google, Amazon and Microsoft investing heavily in AI. As the hunt to acquire the top talent continues, these companies are expanding. For instance, in June 2015 Facebook opened an AI lab in Paris to supplement its New York and Silicon Valley bases allowing Facebook to onboard some of the top talents in AI and machine learning software development that Europe has to offer. That said, according to research recently launched by CognitionX, ‘London: The AI Growth Capital of Europe’, London is home to double the number of AI companies than its closest rivals Paris and Berlin combined.

That is the macro view, but looking across industries, AI adoption is mirroring the general move towards digital transformation. Sectors like healthcare have much to gain from AI investments, with tangible benefits such as accessible healthcare, thanks to doctors’ apps and increased speed and accuracy of diagnosis. The sectors that have already established themselves within the digital realm are best placed to exploit AI.

There is high adoption of AI in telecoms, automotive and financial services. Retail, consumer packaged goods and media are classed as medium adopters, according to a McKinsey report. Perhaps surprisingly, healthcare is in the lower ranks of AI adoption; this is an area ripe for innovation considering the potential human benefits. Meanwhile, one of the key characteristics noted of early AI adopters is the current focus on growth over savings.

What does this mean for the workforce specifically, and what challenges does this throw up for the c-suite looking to integrate AI capabilities without rocking the boat? The first concern to address is that AI will outsmart human capabilities and steal employees’ jobs. Google’s Alpha Go-playing AI beat the grand master in March 2016. And Cambridge Consultants’ ‘Vincent’ has learned to mimic and recreate the art masters’ paintings. But we must not forget that while mimicking and analysing patterns is exceptional, we are not so advanced that the human mind can be replicated and replaced.

Shifting focus

The focus should be on the human benefits provided by AI in the work setting. It will ultimately impact organisations across the value chain from forecasting to optimised production, targeted marketing and enhanced user experiences. By saving time on more admin-heavy tasks, employees can embrace more creative and strategic tasks and, potentially, even enjoy shorter working days, while winning more manageable workloads. Businesses that communicate this well and implement a successful AI strategy will gain the competitive edge.

Technological changes have always caused trepidation. As the workplace has evolved over the decades, we have had to upskill and adapt. But once employees gain a full understanding of the benefits of this technology, the applications and tangible impacts to their day-to-day work, it will be far easier to progress. This will require strong leadership, expert communication and a holistic education for everyone. By taking colleagues through the process together, championing AI and sharing in the experience, employees can take comfort and learn to embrace the transformation.

For those averse to upskilling and retraining, the perspective needs to be switched. With growing streams of data, understanding how best to utilise the data and use it to a business’s advantage will be essential, and may require upskilling. To become a truly data-centric business, cultural change is not only inevitable but necessary. This can provide new and exciting opportunities for the workforce.

But for some this might not be convincing. Workforces will flex to meet the needs of a collaborative AI workforce if they trust in the technology. As we have seen with the implementation of the General Data Protection Regulation, trust in the digital sphere requires compliancy and transparency. Governments and businesses are working to ascertain the best way to regulate AI without pouring cold water over its potential for exponential growth, potentially putting businesses at a disadvantage.

Gradual impact

Enter, explainable AI. 67 per cent of business leaders hold the belief that AI and automation will have a negative impact on stakeholder trust levels in their industry within the next five years. This means that work is yet to be done. Explainable AI or XAI describes a system where the actions of an AI can be easily understood by humans. By helping to understand how and why an AI came to a certain decision, it creates more transparency. As we begin to regulate the digital sphere, in order to foster consumer trust, reinforce data privacy and prepare for a brighter, data-driven future, it is clear that Explainable AI is the way forward. It will serve to both alleviate the concerns of the c-suite and enable them to apply a soothing balm to those of their employees. In healthcare and mobility services, this will be particularly critical.

However, challenges for Explainable AI include the unveiling of business-critical algorithms: essentially, a business’s intellectual property. This is anti-competitive, as is its trade off with performance vs transparency. As explorations continue, basic administrative AI applications are rolling out, from transcription services to chat bots and predictive email text.

The real question has to be not ‘when will they figure this out so we can get on with our AI integration’, but ‘how can we prepare our workforce for a data-driven business strategy optimised with AI?’ As with most so-called technology revolutions, the impact is gradual. With the pace of change accelerating however, there is no time to rest on your laurels. Complacent businesses will be left behind. Cultural change is the first step in preparing the workforce for an AI-led future – so start now.

Matt Watts, Director, Technology and Strategy, NetApp
Shutterstock/Mopic