AI Reskilling: Embrace the Change or Fall Behind

An opinion piece by the VP head of Northern Europe at Databricks

Michael Green, VP Head of Northern Europe, Databricks

December 21, 2023

4 Min Read
Illustration of a business man in a suit
Getty Images

Generative AI has the potential to totally revolutionize businesses all across the world. Its ability to increase efficiencies and reduce costs is set to raise global GDP by $7 trillion (£5.75 trillion) over the next 10 years. This recent explosion of generative AI has brought into focus just how important it is for businesses to take steps to incorporate AI into more aspects of their business. If they do not, they risk being left behind and miss out on this generational opportunity.

However, for such a transformative technology to be fully adopted, it is crucial for workforces to be adequately equipped with the knowledge and skills to harness it. There is no time like the present and consequently organizations must prioritize the development of an upskilling strategy to ensure that their teams have the tools to thrive in this rapidly changing environment.

Embracing the AI revolution

With parallels to the industrial revolution of the 18th century, the data and AI revolution is now upon us. As with all moments of rapid technological advancement, rates of adoption can vary, with some organizations being more risk averse or resistant to change. There are some who are apprehensive about implementing AI into their businesses, however, they increasingly find themselves in the minority. An overwhelming majority (88%) of those surveyed in a recent MIT report said they were using generative AI, with 26% investing and adopting it, and 62% in the experimentation phase.

This full embrace of the AI revolution is the first step, but to back it up, it is crucial for businesses to have robust data foundations in place. Many businesses are finding that even though the desire is there, they are bogged down by legacy data architecture and processes that are not fit for purpose. Data is the lifeblood of AI, and having a system in place to manage it properly is therefore of paramount importance before embarking on more ambitious AI projects.

Opening up generative AI to the whole workforce

How then can businesses go about incorporating this approach into a comprehensive and coherent strategy? As with most significant business efforts, the answer lies with its people. The same MIT report found that 41% of respondents in EMEA thought training and upskilling was the biggest pain point to overcome in generative AI adoption. By focusing efforts on ensuring that people across the organization have the correct skills to be able to leverage the transformative effects of generative AI, businesses will put themselves in the best position to reap its benefits.

Clearly, people who are already in technical roles within their organizations − such as data engineers, analysts or IT experts − must be a focus of any wider upskilling effort. Often this part of the business is responsible for any sort of technological implementation and therefore it is crucial that the strategy prioritizes a culture of regular learning to ensure that these technical workers are kept up-to-date with a constant and ever-changing AI landscape.

Yet, for organizations to fully realize the wide-reaching potential of generative AI, they must expand any upskilling efforts beyond their technical workforce. Generative AI lets those who do not necessarily have a technical background to nonetheless be able to embrace its opportunities. For example, workers from all parts of the business can benefit from generative AI’s ability to break information down into easy-to-understand language, providing a massive boost to training efforts, or assistance with a wide range of repetitive writing tasks. All of this can improve efficiencies, reduce costs and drastically improve decision-making processes.

Building foundational skillsets

One of the most popular ways forward is the use of large language models (LLMs), such as the one used by OpenAI’s ChatGPT, or Google’s Bard. However, these LLMs have their drawbacks – there are issues around cost and environmental impact, but most importantly, inaccuracies. As these models take information from all over the internet, and learn from people’s use, their data has the potential to become unreliable. When AI is trained on untrustworthy data that is not scanned for quality, accuracy or bias – it could lead to disastrous results that may damage the reputation of a business.

A solution to this challenge that is becoming much more popular is creating a smaller language model in-house, instead of relying on large models that pull information from disparate sources. If an organization has the correct skills in place, it can create a smaller language model, involving much fewer but more tailored datasets. By building in-house, companies can benefit from greater control over their sensitive data, realize much lower costs when it comes to deployment, as well as highly accurate outputs and improved sustainability.

AI will be the most important technology in the decades to come, and its potential to improve outcomes for businesses everywhere is almost limitless. Therefore, firms must ensure that their people are set up with the correct skills to be able to fully maximize this opportunity.

About the Author

Michael Green

VP Head of Northern Europe, Databricks

Michael Green is the VP Head of Northern Europe at Databricks.

Keep up with the ever-evolving AI landscape
Unlock exclusive AI content by subscribing to our newsletter!!

You May Also Like