British-Hungarian philosopher Michael Polanyi aptly summarised his theory on human knowledge and intelligence in seven words: "We know more than we can tell." His theory, later called Polanyi's Paradox, suggests that humans possess a vast amount of knowledge on how to perform tasks that we cannot explicitly explain, like face recognition or driving a car in traffic. You can't really put your finger on it or compose the right words to describe it, but you just know that you know!
Long before today's workforce started to worry about getting replaced by robots and machines, Polanyi was confident that human knowledge was 'beyond our explicit understanding' and could not be codified.
Or is it? Advancements in generative artificial intelligence (AI) are showing signs that tacit knowledge is fast emerging as an opportunity. While the current version of ChatGPT already analyses explicit knowledge and structured (quantitative) data, the multimodal Chat GPT-4 (currently only available to subscribers) is also able to make sense of tacit knowledge, provided it's been digitised.
Impressing early users, Chat GPT-4 can extract insights and patterns from large collections of unstructured data. This includes sources such as email, images, videos and instant messages. In addition to simplifying coding, it can rapidly create a website and even pass exams with flying colours.
So, if generative AI can successfully translate tacit knowledge to explicit, what impacts and opportunities will this create for knowledge workers?
The brains of an organisation
Tacit knowledge is one of a company's most valuable intangible assets, besides goodwill, brand recognition, and intellectual property rights. Intangible assets account for 90 per cent of the value of all S&P 500 companies.
Tacit knowledge is the intelligence, wisdom, intuition and insight people gain through their personal and professional experience. It's not codified or written down and is difficult to put into words, rendering it a risk for every organisation. Why? Because every time a person leaves an organisation, their tacit knowledge leaves with them.
Google for example, has acquired more than 200 companies in the quest to expand its internet empire to different industries. Imagine how much tacit knowledge has been accumulated through multiple acquisitions and over two decades in the business.
The ability to capture who and what your business is – all the information, employee experiences, contexts and background stories of every decision made within the organisation from the very beginning – is beyond powerful to both build insights and fast-track solutions. Imagine leading a team or an entire organisation with the help of an AI who knows the ins and outs of your business history more than you do. The ability to brainstorm with it, collaborate, debate, and make important decisions.
Cleaning up the mess
Engineering has often been described as messy work as it draws on both explicit and tacit knowledge, together with both structured and unstructured (quantitative and qualitative) data sources, which include spreadsheets and databases, reports, drawings, plans, and emails.
The 'messiness' of engineering has historically presented organisational challenges since tacit knowledge is difficult to transfer, and the analysis of large volumes of unstructured data has been beyond the realm of possibility – until now.
Technological advancements are driving many companies to build or customise proprietary, domain-specific generative AI models that are trained with their own data and for their own purpose. By turning the technology inwards, these companies are rapidly learning how to analyse their unstructured data and encode tacit knowledge to enable better insights and intelligence, ultimately gaining a competitive advantage.
Multinationals such as Bosch and Siemens are creating AI-based solutions to capture the tacit knowledge of their engineers and technicians. By analysing data from manuals and repair logs, they are extracting patterns and insights that can be used to improve the efficiency and effectiveness of maintenance and repair processes.
NASA is using generative AI to capture and share the expertise of its space exploration workers. It's analysing data from images, videos, and mission logs to improve the design and performance of spacecraft. Haleon (previously GSK) as well as Saatchi & Saatchi are using the AI-powered search engine Lucy to 'liberate corporate knowledge' and break down knowledge silos. Others, like PepsiCo, are using Starmind to make tacit knowledge sharing and collaboration easier, fast-track innovation and decision making, while enabling contributors to gain recognition by showcasing their expertise.
And a growing list of companies, including IBM, are using generative AI to capture, analyse, and share data from emails, chat logs, and customer feedback to extract insights and patterns that can be used to improve customer engagement and sales performance.
Smart people to make smarter decisions
You can't outsource great ideas to AI, but it can definitely help you develop them.
The ability of generative AI to not merely unlock but preserve tacit knowledge will, in the words of management consultants Arthur D Little, "allow smart, experienced people to make even smarter decisions". It will equip knowledge workers with greater cognitive capacity to do and engage more with the problem they are trying to solve, providing the ability to make sense of mountains of info and data. Librarians could move from schools to offices as the importance of tagging, storing and understanding how data should be collected and structured will increase.
Of course, as with any other technological advancements, the rise of generative AI is accompanied by cries of negativity side by side with the opportunity. According to Goldman Sachs' economists, 300 million jobs could be automated by AI platforms like ChatGPT and 18 per cent of work globally could be computerised, all of which could lead to job losses. This would probably only be a short-term impact since innovations such as this have historically created more jobs in the long run – but only if we embrace it.
The key for organisations is to get ahead and be a leader in this space, otherwise those who can't and won't compete will eventually lose. If we see AI only as a threat, it will never be our ally.
Knowledge workers will always remain valuable in roles that require critical thinking, creativity and human interaction. Skills such as decision-making, problem-solving and stakeholder engagement require human judgment and experience, and are difficult to automate.
The winners will be the organisations that use generative AI to augment, not replace, humans to more expeditiously make more informed decisions, solve problems and innovate.