Where do you turn for guidance?
Our business is (and we suspect that yours might also be) a lot like AI, which bears a striking resemblance to pre-internet knowledge sharing, but faster, more pervasive, and with access to whatever it (AI) can get its hands on.
We asked ChatGPT to summarize this and ended up with the following structure and sequence (Note: we've spared a lot of detail, but I'm sure you'll get the idea).
Pre-Internet Knowledge Sharing:
1) Oral Tradition: For centuries, before the invention of writing, knowledge was shared orally. While rich in context, it was prone to distortion over time with each retelling.
2) Written Records:
- Invention of Writing (Circa 3200 BCE): The development of writing systems allowed for the recording of knowledge on clay tablets, papyrus, and later on paper. Written texts became the primary means of preserving and sharing knowledge.
- Manuscripts and Books: Before the printing press, manuscripts were laboriously hand-copied by scribes, making books rare and expensive.
- The Printing Revolution (15th Century): The invention of the printing press democratized knowledge, enabling the mass production of books, accelerating the spread of knowledge across Europe and later the world.
3) Institutional Learning:
- Universities and Academies: With the establishment of universities in the Middle Ages, institutions became key centers for knowledge sharing.
4) Industrial Era and Knowledge Distribution:
- Newspapers and Journals (17th Century Onward): The rise of newspapers and academic journals allowed for the regular distribution of information, including scientific discoveries and public discourse, but still limited by geography, literacy, and social status.
- Libraries and Archives: Public libraries became vital repositories of knowledge, providing access to books and records. However, physical limitations still constrained the speed and reach of knowledge sharing.
5) Digital Revolution: We'll let you fill in the blanks here.
Today...
6) Artificial Intelligence: What limitations could it possibly have? AI's Self-Assessment noted the following:
- Accuracy: Data Quality Issues, Bias in Algorithms, (lack of) Contextual Understanding.
- Completeness: Data Silos, Limited Understanding of Human Factors, Narrow Focus.
- Timeliness: Dependence on Historical Data, Lag in Data Processing, Inflexibility in Adapting to New Data.
- Examples:
a) Strategic Decision-Making: AI can assist in making strategic decisions by analyzing trends and projecting future scenarios. However, if the data does not account for sudden changes in the market or emerging trends, the AI's recommendations might lack accuracy and timeliness, potentially leading to misguided strategic choices.
b) Talent Management: AI-driven HR tools can help in hiring and talent management, but they might overlook qualitative aspects like cultural fit or leadership potential. Furthermore, if the AI system is not regularly updated with the latest data on employee performance or industry standards, its recommendations might be incomplete or untimely.
So, how do we mitigate these limitations?
AI tells us, do this:
-
Data Governance: Implement strong data governance practices to ensure the data fed into AI systems is accurate, complete, and up-to-date.
-
Human Oversight: Maintain a balance between AI-driven insights and human judgment, especially in areas where qualitative factors are critical.
-
Integrative Approaches: Use AI in conjunction with other tools and methodologies to ensure a more holistic approach to business management, capturing both quantitative and qualitative aspects.
What does this really mean?
Be careful where you get your information. Ask around and inquire whether there are better sources (other experts in that domain). Test their knowledge. Do they actually know something (have an insight) that you might not know? Do they have evidence? Can they share that evidence in a compelling manner that adjusts your understanding? Do they know what to do with their findings? Anybody can find problems, but do they know 'what next'? How do we use that information to move forward in a satisfactory manner? Are they willing to admit when they do not have experience and don't know the full answer? Will they guide you to a better source of truth (i.e., someone with more subject matter expertise)?
