Maximising Human-AI Collaboration to Elevate HR Impact
In today’s digital age, where AI adoption has become a competitive necessity, the question facing every HR leader is not whether to include AI in their strategic toolkit but how best to integrate it.
AI integration will only be as effective as its collaboration with the end user—humans—a notion underscored by recent research at MIT. Studying the synergy between AI and radiologists, the research revealed that while AI systems can rival or surpass human abilities in specific tasks, human biases in integrating AI insights can significantly undercut the potential benefits of such collaboration.
This sentiment is echoed by Eric Siegal', founder of Machine Learning Week, during his discussion with David Green on the Digital HR Leaders podcast. Further emphasising this point, he shared,
"IBM recently came out with research showing that according to executives, the average returns on AI projects is lower than the cost of capital. So, basically, no returns, and I believe that's largely because they failed to deploy".
As we delve deeper into the world of AI, it becomes increasingly apparent that while technology is advancing rapidly, humans are still an integral part of its success. Therefore, as we seek to maximise the potential of human-AI collaboration, it's critical to address and recalibrate our strategies to harness its full potential. This requires a shift in mindset from simply using technology as a tool to collaborating with it as an equal partner.
Challenges in Human-AI Collaboration
While neither humans nor AI systems are perfect, researchers examining the human-AI partnership have identified several challenges that could hinder the realisation of its full potential. These challenges include:
Lack of Trust
As Bernard Marr, the author of ‘Data-Driven HR: How to Use AI, Analytics and Data to Drive Performance,’ insightfully remarked on the Digital HR Leaders podcast,
"I think lots of HR teams are scarred from previous automation experiences where they try to install a chatbot, for example, to help answer HR questions, and previous AIs were just not up to the task."
Consequently, despite the advancements made in AI technology, there is still lingering scepticism about its capabilities. Why should employees trust AI insights when there have been instances of chatbots providing incorrect information and causing confusion? This lack of trust can hinder humans' willingness to collaborate with AI and, in turn, limit the potential benefits of such a partnership.
Explainability
Rooted by the "black box" problem, AI, such as predictive analytics in HR, is perceived by many as unexplainable. However, just as you would approach any partnership, meaningful collaboration cannot be predicated on blind trust—especially in critical decision-making scenarios, where understanding the reasoning behind AI-driven decisions is crucial.
Ethical Considerations
AI, as it currently stands, can amplify human biases and unfair practices if not employed and monitored with diligence. Consider Amazon's infamous AI-hiring tool, trained on data from the past decade, which led to significant gender bias against women in recruitment.
This raises questions about the ethical implications of human-AI collaboration and the need for organisations to adopt ethical frameworks to guide their partnerships.
Replacement of Jobs
Then we have the fear of replacement of jobs. AI will change the nature of work. It is inevitable and already underway.
As such, some employees and even organisations have chosen to limit the role of AI in their operations as a safety response in a bid to save jobs. However, this approach is counterproductive. It stifles innovation and prevents organisations from fully harnessing the potential of human-AI collaboration.
Resistance to Change
Lastly, the resistance to change. When AI is introduced into an organisation, it brings with it changes in processes, roles, and responsibilities. These changes can be met with resistance from employees who are not willing to adapt to the new ways of working—especially when this is met with fear of job loss, ethical considerations, and lack of trust in its capabilities.
Strategies for Overcoming Collaboration Hurdles
Overcoming the above mentioned challenges requires a multifaceted approach involving technical, organisational and cultural considerations. Below are some strategies for organisations seeking to optimise their human-AI collaboration:
Understand the Core of Resistance
Whether it be trust issues, fear of job displacement, or simply reluctance to embrace change, the common theme is resistance. Therefore, understanding the root cause of this lack of positive employee experience is critical for successful collaboration.
Once you have a good idea of what is causing resistance, conduct pilot projects and engage with employees to address their concerns while demonstrating the benefits of collaboration.
Advocate for a culture of open communication where employees and management can freely express their thoughts, concerns and ideas about working with a machine. Ensure you listen to these concerns and openly address them at every opportunity.
Change Management
"People aren't generally applying change management techniques for machine learning, and that's why they're failing, because they're so focused on the core technology like, "This is the best technology, surely it's going to be valuable". But again, the purpose of the project isn't to do awesome, cool number-crunching, as much as I wish it were, it's to make use of that to actually improve an operation, which means change.",
Eric Siegal shared during his interview with David Green.
As Eric opined, AI integration into an organisation is change management at its core and should be treated as such.
Invest in AI that Matters
"The really important point is to figure out which bits AIs can do very well and people are happy with, and where do we need people.” quoted Bernard Marr.
In other words, invest in AI that complements your human workforce and employee experience, not replaces or diminishes it.
For instance,
"if we simply outsource our recruitment to an AI, that is dangerous, because we've seen this in companies that have had lots of problems where the AI was biased towards certain genders or races and other things. So, what we need to do is use AI in the right way.
So, AI in recruitment, for example, is really good at finding potential candidates that have skills that you might not have looked at before. They are really good at helping to assess certain skills by running games, for example, something Unilever has been doing for a long time, where they say, "Okay, I want to understand the risk appetite of candidates", and they then create little online games that help people assess or help to understand what your risk appetite is. So, these are really good, but I think at some point you still need the human touch in recruitment, and I think we are a long way away from AIs being able to do all of this." explained Bernard when asked about finding the right balance between AI and human intervention.
Therefore, human-AI collaboration must be strategically planned and executed to maximise the organisation's benefit. Assess your organisation's current state, identify areas where AI can add significant value, and invest accordingly.
However, drawing insights from Mohammad Hossein Jarrahi, Kelly Monahan, and Paul Leonardi in their Harvard Business Review article, there are specific capabilities that your AI needs to acquire in order to work with humans effectively:
NLP (Natural Language Processing): To ensure that AI can effectively understand and respond to human communications.
Explainability: Essential for fostering trust and acceptance among users, investing in explainable AI allows humans to understand how a decision was reached.
Adaptability and Personalisation: AI increases its value to users by learning from past interactions and tailoring responses based on individual user preferences or needs.
Context-Awareness: This enables AI to comprehend and respond, taking into consideration a variety of factors such as time, location, or task urgency.
Upskill your Workforce
Eric Siegal highlighted a poignant truth during his conversation with David Green.
"Upskilling is first and foremost the most important thing."
And he couldn't be more right. These machines will be working and run by humans, and for this to work effectively, organisations must invest in upskilling their workforce.
Referring back to Jarrahi, Monahan, and Leonardi's HBR article, bridging the competency gap in AI-enabled organisations requires a well-thought-out human capital roadmap that invests in:
Data Literacy
This encompasses the ability to identify and assess the relevance and credibility of data, validate results through empirical methods such as A/B testing, and adeptly create and customise visual data representation to communicate findings effectively across varied audiences.
AI Literacy
Understanding the mechanics behind AI—how algorithms are constructed, function, and can be harnessed to augment human decision-making—is critical. This literacy extends to recognising the inherent limitations and potential biases within algorithmic processes, ensuring a more informed and critical application of AI tools.
Algorithmic Communication
The art of algorithmic communication—instructing algorithms through precise and thoughtful input (e.g., prompt engineering) and interpreting their outputs for diverse stakeholders.
As we stride toward an increasingly automated future, the amalgamation of data literacy, AI literacy, and algorithmic communication forms the triad of essential skills for surviving and thriving alongside artificial intelligence. This upskilling pathway prepares the current workforce for the demands of tomorrow but also ensures that human-AI collaboration adds tangible value to the organisational ecosystem.
But we need to encapsulate humility and curiosity mindsets to do this effectively.
Referring back to David Green and Bernard Marr’s conversation, Barnard eloquently pointed out,
"If you have enough humility to say, 'Actually, I don't understand everything and I want to learn', and the curiosity to then do this, that then pushes you towards this continuous learning mindset. I think those are the key skills that any HR team and any HR leader needs to foster. Realise actually I don't know everything, there's so much happening out there, this is going to change our department, it's going to change how we work, and I need to continuously learn and develop to keep up-to-date with this."
He continued, "The competition for your jobs in the future will not be AIs, it will be humans that use AI really effectively".
Thus, it becomes imperative for organisations to cultivate an environment that leverages powerful technology and significantly enhances it with human skills, paving the way for the creation of superhuman capabilities.
The integration of AI into our workforce is not a matter of if but when. And as HR Leaders, we have the onus to ensure that we prepare ourselves and our teams for this impending revolution.
Through a deliberate effort to overcome resistance, leverage the right technology, and invest in human potential, HR and people analytics leaders can chart a course towards a more collaborative and prosperous future.
Unlock the potential of your people analytics function with our Insight222 People Analytics Program®
At Insight222, our mission is to make organisations better by putting people analytics at the centre of business and upskilling the HR profession. The Insight222 People Analytics Program® is your gateway to a world of knowledge, networking, and growth. Developed exclusively for people analytics leaders and their teams, the program equips you with the tools, insights, and connections you need to create greater impact.
As the landscape of people analytics becomes increasingly complex, with data, technology, and ethical considerations at the forefront, our People Analytics Program brings together over one hundred organisations to collectively address these challenges. Discover how joining the Insight222 People Analytics Program® can help you deliver real business value to your organisation.