Thinking Outside the Bot: How AI Fuels Creativity

Intelligent Knowledge Management

In previous sections, we explored how AI broadly accelerates discovery processes, reshaping the innovative focus of people working with it. In many cases, AI models help us get to decision points faster, where humans then apply creativity to solve problems with real-world impact. While broad AI-driven breakthroughs reshape entire fields, AI’s subtle yet profound influence on daily innovation comes from solving the smaller but frequent hurdles associated with managing and retrieving knowledge. In other words, AI helps us manage tasks which often distract us from meaningful innovation.

AI Overviews

Intelligent knowledge management (IKM) is fundamental to most public use of large language models (LLMs), as users commonly leverage AI to retrieve and distill information, with AI overviews (also called, “AI summaries”) complementing or replacing the ubiquitous Google search results. Such knowledge retrieval becomes particularly powerful when combined with retrieval-augmented generation (RAG), a technique that allows public AI models to respond to targeted, domain-specific questions. For instance, while general-purpose models such as ChatGPT, Claude, or Gemini effectively handle broadly available knowledge, they can’t inherently address detailed queries against internal data sources (“Is my department exceeding its budget for this quarter?”) or other specialized contexts. Techniques like RAG fill this gap, embedding local data directly into the AI query-response interaction, allowing medical researchers to swiftly retrieve specific patient trial data from internal repositories, the engineering manager to understand her departmental budget, or the nonprofit to categorize sensitive donor profiles. In each case, the professional minimizes their time spent on information retrieval, and instead applies their focus on the question of how to innovatively use the resultant information.

Business models surrounding web traffic and AI overviews are yet to be sorted out. The economics of the web are based on site visits, and websites without paywalls typically make money from advertising. If the content from those sites is increasingly scraped, synthesized, and provided via AI overview, web traffic to those sites declines–a trend which has already begun. In turn, revenue for the source sites declines. Mitigating this problem will require creative market incentives, and possibly regulatory action. For this reason, AI overviews represent a risk to the web economy we’ve known for decades, despite their many personal advantages.

Copilots

Another example of AI’s ability to automate IKM is seen with the AI-powered copilots which increasingly attend meetings. These tools don't simply transcribe conversations; instead, they summarize critical discussion points, automatically reference relevant documents, and generate action items for participants. This automation reduces cognitive burdens, saving time from compiling notes and actions, and ultimately freeing individuals to engage more rapidly in strategic and creative activities.

Such copilots also manifest themselves in other applications, each designed to provide users with the right information at the right time. Their capabilities now are relatively rudimentary compared with what is coming: future versions will draw on numerous organizational data sources to provide expert context-sensitive assistance to employees. Given that organizations are inherently inefficient across structures and priorities, particularly across windows of time, future AI copilots will become the bridge helping companies significantly improve outcomes.

Two emerging capabilities may expand the scope of intelligent knowledge management: predictive knowledge-gap detection and tacit knowledge conversion. Although not yet widely used, they promise to further assist us in the workplace, and possibly in our personal lives. Let’s explore each in turn, with examples.

Predictive knowledge-gap detection

Predictive knowledge-gap detection proactively identifies blind spots within organizational teams, highlighting areas where essential expertise, data, or insights may be lacking. By reducing these “unknown unknowns,” teams can manage risk more effectively, avoid costly oversights, and make better strategic decisions. This concept is best illustrated with an example: consider a hospital running an AI-enabled clinical decision support system. In this hospital, certain physicians may be consistently ordering improper tests for a given set of symptoms. While traditional human intervention might spot this pattern sporadically, providing ad hoc suggestions to the doctors, AI often has the ability to see further into the roots of the problem since it draws from numerous data sources. Rather than simply flagging the test misuse, for instance, AI might also observe deeper actionable factors at play. For example, it might note that physicians from particular residency programs are 40% more likely to order these tests, and that trends worsen under certain scenarios, such as when ambiguous symptoms are presented. Extrapolating into the future, these patterns, i.e., this knowledge gap, will drive higher costs, diagnostic delays, and patient dissatisfaction. Armed with this knowledge, relevant education can be targeted to that physician cohort, and onboarding training can be modified to reduce this problem in the future. AI is now inferring knowledge gaps based on patterns of behavior, not on test scores or explicit feedback. This analysis allows both the immediate problems to be addressed, and importantly, the underlying causes for those problems, helping those involved to improve their performance.

Tacit knowledge conversion

Similarly, emerging AI capabilities in tacit knowledge conversion represent a modern evolution from legacy expert systems. Historically, capturing tacit knowledge, e.g., expert insights, intuition, or learned experience, required extensive manual effort and explicit documentation. Modern AI changes this equation, enabling more intuitive and less burdensome knowledge capture. For instance, AI-driven assistants can analyze expert dialogues, extracting and structuring implicit knowledge into searchable repositories, reducing the overhead traditionally required for knowledge preservation and transfer. To illustrate this, consider a large civil engineering firm employing multiple bridge engineers. The senior-most engineers have decades of experience, mostly embedded as tacit knowledge–not in manuals or software–but in their learned experience and subsequent professional behavior. AI systems might analyze historical project data, including annotations and notes, emails and notes exchanged during the design-and-build process, and cost and performance outcomes over time. With this ingested, certain best practices may be extracted and synthesized, for example, “In coastal environments with high salinity and fluctuating freeze-thaw cycles, prestressed concrete girders require a specific type of sealant and expansion joint configuration to prevent hairline cracking.” This capability reduces the complexity and cost associated with manual knowledge capture, and minimizes quality fluctuations when senior people leave the organization. Such tacit-to-explicit knowledge conversion harkens back to expert systems of past decades, but now, knowledge capture is accomplished by neural networks, reducing the cost and complexity of building repositories for step-by-step guides and other institutional knowledge.

Tying the Intelligent Knowledge Management pieces together

AI-based IKM swiftly surfaces precise, contextually relevant information, freeing human innovators from routine information-seeking tasks. With immediate access to refined knowledge, professionals more quickly engage in creative ideation, experimentation, and strategic decision-making. Importantly, these AI-driven solutions will seamlessly integrate into existing productivity and collaboration tools, making their benefits accessible without disrupting established workflows. While challenges remain, such as ensuring accurate, bias-free outputs and safeguarding data privacy, advancements in techniques like RAG and knowledge embedding mitigate these concerns, offering reliable and secure ways to harness organizational knowledge. Looking ahead, AI-based IKM functions will become increasingly personalized and context-aware. As AI evolves to align more closely with individual needs and organizational workflows, boundaries between human ingenuity and machine assistance will blur even further. This convergence will empower individuals and teams to innovate with unprecedented clarity, efficiency, and impact.

About the Author

With a background in artificial intelligence/machine learning (AI/ML), cloud computing, and internet of things (IoT) technologies, Rick Hamilton is a named inventor on more than 1,060 issued US patents, making him one of the most prolific inventors in world history – just behind Thomas Edison. He has more than 30 years of patent portfolio development and governance experience, and 13 years of portfolio usage and organizational strategy experience. This includes establishing and leading patent strategy for a Fortune 10 healthcare company. He has spoken on artificial intelligence/machine learning, innovation and IP management, cloud computing, and IoT technologies in 32 countries, and has trained thousands of technical and business staff on best invention practices.

Rick can be reached at rick@hamiltonandboss.com with questions or comments.

Ready to secure your strategic advantage?

Visit us at hamiltonandboss.com to learn more or schedule a consultation.

Follow us on LinkedIn and X | Contact: info@hamiltonandboss.com