Thinking Outside the Bot: How AI Fuels Creativity

A New Kind of Intelligence in the Workplace

Humanity’s efforts to construct a new kind of intelligence are finally bearing fruit, after decades of false starts and brittle solutions. While the inner workings of this emergent intelligence are different than those processes occurring in the human brain, many of its observable effects are similar to those found with human intelligence. But fear of AI and its impacts misses a critical point: these systems will not replace us en masse, but rather change the way we work, much as past industrialization took our ancestors off the farms and put them into factories. As factory automation increased, subsequent generations became more likely to work in services job and to otherwise work with ideas, designs, and sophisticated systems. Similarly, AI agents will become our assistants, helping us achieve much more than previously possible. The key to our future professional success will lie, in part, in our ability to work with them, leveraging AI to solve problems in effective fashion.

AI as a General Purpose Technology

How might the creation of a new form of intelligence accelerate the rate of human innovation? To answer that question, consider that AI is a general purpose technology, like electricity, although unlike electricity, AI’s effects compound over time. No one today asks a business, “what is your electricity strategy.” Instead, we recognize electricity’s ubiquity across all segments of the economy. Electrical power changes the way we innovate, because it fundamentally mitigates the time and toil associated with tasks ranging from laundry to industrial manufacturing, and it likewise makes new products and services possible—without it, we would not have diverse solutions, ranging from telephones to diagnostic imaging, from modern air travel to television broadcasting, from credit cards to digital gaming, along with countless other products and services.

Similarly, AI—in its current and future forms—opens many new doors, and enables numerous novel products and services. AI will, in many cases, reinforce our most natural tendencies and help us achieve our goals. For those aspiring to make a difference in the world—whether as a scientist, engineer, artist or entertainer, business leader, or governmental official—AI will provide tremendous benefits.

Two Pathways to Innovation: Discovery and Capacity

At a broad level, AI aids innovation through accelerated discovery, as well as expanding human intellectual capacity. By accelerated discovery, we see repeatedly that intelligent and educated people spend time working on “precursor problems,” which do not in themselves provide direct benefit to businesses or humanity, but which are necessary to solve before such an impact can be made. AI has the advantage of accelerating the solution of such precursor problems, so that humans can shift their focus to the point of impact. To understand the expansion of human intellectual capacity, consider that AI offers many techniques to prepare us, on demand, for the next challenge–making us better versions of ourselves. We'll examine both of these effects over a short series of blogs, exploring two examples of accelerated discovery and three ways that AI expands human intellectual capacity, all detailing how AI promises to enrich our ability to innovate.

Rapid Data Analysis and Pattern Recognition

The Power of AI in Finding Hidden Insights

To begin a discussion on accelerated discovery, consider that, at its simplest level, AI can process vast amounts of data quickly, and excels at detecting hidden relationships that humans might miss. These capabilities, when applied with insight and experience, help organizations uncover breakthroughs faster and make better decisions across industries. Time and time again, AI accelerates knowledge tasks by some varying percentage, perhaps 50 to 99%, leaving the last mile, high value-add decisions to humans. This means that frequently the tedious work before creative decision making and strategic assessment can now be significantly compressed in time and effort. We will see this effect at both a macroscopic level as well as in our own professional lives.

AlphaFold and the Protein-Folding Breakthrough

One of the most publicized examples of AI performing exceptional data analysis was the Nobel Prize-winning solution for the protein-folding problem, through the efforts of Google DeepMind. The responsible model, AlphaFold, has now iterated multiple generations and inspired spin-offs, but to understand its value, let’s look deeper at the problem, before and after this creative application of artificial intelligence.

For more than half a century, biologists wrestled with the protein-folding problem—the quest to predict a protein’s three-dimensional shape from its amino acid sequence. In principle, this sequence dictates the protein’s structure, yet a single protein could theoretically take on an unimaginable 10^300 possible conformations. Decades ago, American molecular biologist Cyrus Levinthal highlighted this paradox: exhaustive search is hopeless, even though real proteins fold within seconds. In other words, if a protein were to try every possible shape sequentially (exhaustive search) to find a stable, functional structure, it would theoretically take an astronomically long time to fold (longer than the age of the universe, in many cases). However, in reality, proteins fold into their functional structures within seconds or even milliseconds. Levinthal used this to argue that proteins cannot possibly fold through trial and error, but instead followed efficient, directed pathways. But could we collectively predict these pathways?

Before artificial intelligence arrived, structural biology advanced only through slow, heroic experiments. X-ray crystallography demanded months or years of painstaking work to coax crystals, collect diffraction data, and interpret electron-density maps. Nuclear magnetic resonance spectroscopy offered a way to study proteins in solution, but only if they were small, and it still consumed vast amounts of time, expertise, and money. Entire teams often devoted careers to solving a handful of structures, and resolving one protein often comprised a PhD dissertation-level effort.

The advent of deep-learning systems such as AlphaFold (and RoseTTAFold) changed everything. These models routinely achieve experimental-level accuracy in minutes, transforming what once took years into an almost routine calculation. Drug discovery, previously a drawn-out and costly process, can now begin with precise structural blueprints, accelerating the design of molecules that hit disease targets. Misfolding disorders - including suspected neurodegenerative diseases - have become more approachable as researchers can finally visualize the toxic shapes at their core. Enzyme engineering for green chemistry, biofuels, and medical diagnostics is speeding up too, because scientists can test “what-if” mutations in silico before stepping into the lab. Perhaps most importantly, the collapse of the protein-folding barrier signals that AI can crack scientific problems once dismissed as intractable, hinting at future breakthroughs in many other fields.

From Mapping to Designing Proteins

With this advance, DeepMind CEO Demis Hassabis estimated–with defensible grandiosity–that AlphaFold has saved roughly one billion years of PhD-level structural biology research time. With folding largely automated, scientists are redeploying–not abandoning–their expertise. Previously, scientists focused primarily on determining the exact shape of proteins—because their shape determines how they function in biology. But now, the emphasis has shifted. Instead of just understanding how proteins look, researchers are actively designing proteins to perform specific tasks or functions. Essentially, we've moved from mapping proteins to deliberately engineering them, moving our focus closer to societal impact.

Investors have followed, as capital has flowed to match the new research velocity. Alphabet-backed Isomorphic Labs raised $600 million to build an AI-first drug-design engine, while AbbVie signed a potential $2 billion option-to-license deal with Gilgamesh Pharmaceuticals that leans on AI-guided molecular design. These headlines reflect an emerging innovation economy in which structure-aware algorithms are embedded from day one of therapeutic development programs.

Critically, human judgment and innovation are more important than ever. AlphaFold and similar models have offered hypotheses that still need experimental validation, cannot see many post-translational tweaks, and struggle with proteins that can take on multiple shapes. Deciding which targets matter, interpreting models in biological context, and inventing the next generation of hybrid AI-lab workflows remain distinctly human problems. The broader lesson is that when AI removes a brute-force bottleneck, our innovative focus shifts to framing better questions and weaving disparate insights into practical solutions—an area where people do their most creative work.

Extending the Model: AI in Materials Science

Perhaps the cynical view is that such breakthroughs are black swans, and that only deep-pocketed technology companies like Alphabet can achieve such results. But AI is performing analogous roles in numerous industries. Consider, for example, materials science, where AI is being applied across a diverse range of industries. Materials science may sound abstract, but it is quickly becoming one of the most practical, results-driven arenas for artificial intelligence. Start-ups such as Citrine Informatics, Noble AI, CuspAI, and Orbital Materials—and mature players like Microsoft and IBM—are all using AI to accelerate the hunt for better batteries, semiconductors, and sustainable materials. Their motivation is straightforward: the traditional “mix-and-measure” approach can take decades, because researchers have had to synthesize and test one candidate at a time. AI flips this model on its head, by mining huge chemical databases and predicting the most promising formulas before a single beaker is lifted.

The payoff is already taking shape. A Microsoft–Pacific Northwest National Laboratory project deployed an AI system to sift through 32 million battery compounds, and home in on eighteen promising candidates in under a week—work that could have occupied a human career. The benefits are practical–one of the resulting cathode materials cuts lithium use, easing a major electric vehicle supply-chain choke point. Similar methods are yielding lighter aerospace alloys, cooler-running power-device semiconductors, and sorbents that trap CO₂ far more efficiently than today’s solutions. These are not simply lab curiosities, as several materials are moving into pilot production or early commercial use.

Pairing AI Speed with Human Judgment

Yet again, the story is not “AI replaces scientists.” Such models still need humans to frame the right questions, judge whether a predicted structure can actually be synthesized at scale, and balance performance with cost, safety, and manufacturability. The most successful programs pair data scientists with materials chemists, process engineers, and business strategists—people who can translate a statistical pattern into a factory-ready product or a new business line. In summary, AI supplies speed and breadth; human creativity supplies judgment and vision. The combination will redefine how fast breakthrough materials reach the real world—and reminds us that innovation is still very much a human enterprise, even as our focus and time are adjusted to topics closer to societal impact points.

A Personal Case Study: AI in Intellectual Property Analysis

Bringing these examples closer to home, consider work which the author has accomplished in understanding competitive business strategies, technical workings, and revenue sources, while helping clients enforce their intellectual property rights. In past years, web-based data gathering and synthesis for complex systems could take many weeks or months, depending on the areas of interest. Despite market intelligence services, understanding and documenting how a competitive solution works, where that product is manufactured and sold, and the revenues associated with it in various jurisdictions was an intensely time-consuming (and therefore, expensive) effort which only laid the groundwork for clients, without proving infringement. Now, with the help of AI, such tasks often take days instead of weeks, allowing faster understanding of the relationship between intellectual property and marketplace solutions, and in turn, more rapid business decisions can be made. Despite certain vendor claims, AI tools cannot generally provide a turnkey, start-to-finish comparison, for instance, of patent claims to competitive products. But they can automate the drudgery and legwork of data gathering and synthesis, cutting analysis times significantly from just a few years ago.

Shifting Human Focus Toward Higher-Order Creativity

Does the use of AI in this case eliminate the need for human innovation? Of course not. Rather, AI has accelerated our interpretation of marketplace ecosystems, allowing our higher-order creativity to be applied to the resultant data set. Work which might have taken dedicated analysts many weeks may now comprise well-crafted prompts delivered to multiple research models, and after a short lunch break, multiple analyses are awaiting our assessment. The human task has shifted from the relatively low-innovation work of data gathering to the critical thought processes of assessing each models’ outputs, synthesizing the most credible elements, and delivering rapid value to clients. The amount of time spent on innovative matters has increased, not decreased!

In the next piece in this series, we'll explore other ways that AI is accelerating discovery, and outline how this is resulting in societal and business gains. In the meantime, consider how you will shift your focus, and accomplish more with these tools tomorrow than you could yesterday.

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