Switching to automatic: how generative AI is impacting corporate L&D

It’s fair to say that ChatGPT has been hitting the headlines recently. Just a few months after its launch, this shape-shifting chatbot from San Francisco-based tech firm OpenAI has already bulldozed its way into the record books to become the fastest-growing internet service of all time – thanks to its 100-million-and-counting global users.

And, with a reputed $10 billion deal already on the table courtesy of Microsoft and a new iteration – GPT-4 – just announced, it seems as if the story is only just unfolding.

Anyone who’s dabbled with the app will understand why it’s prompting excitement and anxiety in equal measure. While we’re used to tech that’s designed to automate repetitive tasks – and, probably, accepting of the inevitability that many manual and mechanical jobs will be lost along the way – these new AI-enabled tools are upping the stakes dramatically, leveraging deep learning and natural language processing (NLP) to perform the kind of knowledge-based tasks that have traditionally been the exclusive preserve of humans.

But, although this encroachment into higher-order functions is ruffling professional feathers all over the world (witness the furore over a Colombia judge’s use of ChatGPT to help make his ruling on a recent court case), there’s little doubt that it’s also opening up fresh opportunities for further corporate disruption and innovation.

Taking personalised learning into the mainstream

For instance, generative AI could have a pivotal role to play in streamlining corporate learning and development (L&D) – automatically spotting missing elements in existing training provision or identifying areas for improvement, for example. Large language AI models like ChatGPT can – in theory – make the acquisition of new skills easier, faster and more intuitive at every level, which is good news for rinse-and-repeat workforce training programmes.

In fact, we’re predicting some quick wins for routine onboarding and skills-based training: using AI-generated imagery from Midjourney or video from Recut.io could deliver a handy shortcut when creating content to help employees learn complex technical procedures. Chatbots could also help writers with initial ideas generation; in an L&D context this might involve posing a question about the components that should be included in a leadership development programme, for example.

The tools’ ability to generate tailored content could also help more precisely tailor L&D programmes, adapting provision to suit individual learning styles and making learning more inclusive.

Moreover, using intelligent automation in this way allows L&D functions to overcome the challenges presented by a more geographically distributed workforce, implementing programmes with greater time and cost efficiencies and enabling organisations to focus the deployment of human resources where they’re most needed – such as engaging in targeted one-to-ones or establishing mentoring relationships.

Perhaps the biggest potential gains, however, are in greater personalisation of the learning journey. If organisations can effectively harness AI to create a bespoke package for each individual employee, they could create an opportunity for L&D leaders to champion a continuous learning loop that’s closely (and dynamically) aligned to their corporate goals.

The human factor

Nevertheless, there’s a balance to be struck between the convenience of smart automation and the necessity for continued human oversight. While tools like ChatGPT may offer many opportunities for the rapid evolution of corporate L&D, they’re not a magic bullet. Humans are still better at conceptual connections, critical thinking and strategic decision-making – particularly when it comes to formulating bespoke L&D strategies for individual organisations.

And, although AI has the potential to optimise time- and resource-heavy processes such as recruitment and onboarding, there are still many important issues to address – not least how to recognise and counter the inevitable systemic biases that are a corollary of the vast datasets used to train generative AI apps.

This shift to greater automation will also create its own need for reskilling and upskilling, as employees begin to manage the concomitant risks and opportunities. Leaders, too, will come under extra pressure as they attempt to confidently steer their companies through change while prioritising the resilience that’s proving to be so crucial to long-term success in uncertain times. Ironically, it’s likely that – in the short term at least – adopting and adapting generative AI tech will increase, rather than lessen, L&D’s departmental workload.

Finding the right way forward will require an enlightened approach – one that’s underpinned by experience and guided by a robust framework designed to facilitate new learning opportunities, however they are manifested. And, while the relentless advance of technology will undoubtedly render some roles obsolete in this brave new world, others will inevitably emerge, providing scope for individuals to become even more creative and productive.

‘I’m really excited to see how AI might be able to improve psychometric testing accuracy, and to use actual behaviour as data sets - there are now tools which can analyse a video of a selection interview and report on a candidate's word choice and the context in which they use them. Early signs are that these tools could play a significant role in removing unconscious bias from hiring decisions, though of course they raise new concerns around privacy and the risk of algorithmic bias. Also, we owe it to candidates to use the AI analysis for their benefit, to suggest development and to help them apply for roles where their talents are a better fit.

To me, the biggest question is what we are going to do with the insights that AI can bring and the time it can save us? If we can invest that time in fostering authentic working relationships, increasing trust and collaboration, and helping everyone achieve their potential, that will be truly transformative.’

Jim Butler – qpeople Senior Consultant