In 1998, two years before the carnage of the dot-com bust, two Stanford Ph.D. students presented an interesting scientific paper. It contained a big idea that made their company one of the most valuable in the world. Their name was, obviously, Sergey Brin and Lawrence Page, and their company one of the then-many search engines — the one that finally gained one of the most lucrative competitive positions in history. The world has since been entranced with their superior digital prowess, and especially their mastery of artificial intelligence (AI) — together with the mound that would create to protect their market position and their valuation. Senior leaders and consultants have studied and attempted to emulate those strategies and practices — with, to date, comparatively scant results.

There’s however another part of that story, lost in the mainstream AI hype, that could help “the rest of us” harness some of the same power. Let’s go back to 1998.

The other part of Google’s AI

Google didn’t invent the search engine but did achieve two things that changed the world. First, its queries became more relevant through a new way of ranking the results’ pages. They did so by measuring those pages’ “centrality” in a broad network of websites, the web pages created by thousands (and later millions) of people that reference each other through hyperlinks. Second, they were quick to embrace and transform the new generation of online advertising, which relied on those queries’ relevance to target users more accurately and, as a result, more valuably. The combination of the two was one of the most powerful business model innovations of the last century.

What most of us don’t realize however, is that Google’s power comes from masterfully using (“organizing”, in Google’s parlance) the fruits of the intelligence of billions of people: the knowledge they create, and the choices they make when they browse. That is, Google’s business model wouldn’t exist without the intelligence of knowledge producers, curators, and users — harnessed in ways unthinkable just twenty years ago, and continuously growing (today, almost 2 billion websites exist). In so doing, Google also contributes to that intelligence, by enabling the world to retrieve knowledge in a sort of collective, comparatively frictionless “remembering”.

The explosion of other social media, from blogging platforms like WordPress to Snapchat, fueled the “creative fire” of millions of people. While much of that creation is of dubious intellectual value, interesting ideas often stem from these environments. And these technologies are increasingly able to collectively “sense” the environment: from the Arab Spring to breaking news and crises, a massively decentralized network of sensors (the majority being simply people with smartphones) has emerged, with its flow of ever-fresh information.

Despite the hype, many internet giants’ strengths don’t just come from the intelligence of their AI, but also from their use of the cognitive power of billions of people who generate information and decide to listen to one another in very specific ways. Trillions of microevents, a sea of new information — all made by human choices — every day. (To us, individually, those choices aren’t worth much, but they make a huge difference to an advertising machine and are worth trillions of dollars of market value.)

The third neural net

These examples matter to most of us. We can take inspiration from them to more deliberately leverage the networks of intelligence that surround us — within and outside of our organizations.

Look at the picture below. Neural-like networks like those, enabled by AI, now span vast numbers of sources of knowledge, especially people but also machines. They weave those nodes together and spread their ideas thanks to web-enabled hyper-connectivity, generation of sensor-based data (from weather to stock inventory, to citizen’s warnings on Twitter), curation of content, display of that content to relevant parties through prediction (think social media choosing what you’ll like to see), and connections (web publishing, synchronous communications, etc.).

Within organizations, that’s an explosion of knowledge available, absorbed and filtered by organizational networks now solidly wired through the likes of Outlook and Slack. That knowledge is then amplified and evolved by social networks and is simultaneously immersed in — meshed with — continuous streams of other ideas curated by AI-based algorithms. Not coincidentally, one of the fastest-growing spaces in the field of analytics is that of “knowledge graphs”, whose biggest advantage is to document and process relationships similar to the ones displayed above.

That’s here, today. That’s why intelligent networks, made of large numbers of people and AI-powered machines, could be a new organizational design ready for widespread adoption. They can help many leaders, from CEOs to middle managers, from centers of excellence to movement organizers, harness the full collective cognitive power of their organizations — to generate and implement stronger ideas, and adapt more swiftly and effectively to fast-changing conditions.

Consider the below: what happens when we add to the traditional management methods (in figure 1, 2, and 3) the ability to deliberately orchestrate networks that span across organizational boundaries (in figure 4)?

Thanks to AI, that can be done today. What could our organizations become by fully using people-machine networks made more intelligent by AI’s “four C’s”, i.e. its ability to exponentially improve the following four things?

1. Connect entities (people, and machines — by, for example, helping pinpoint the right nodes in the network and making them discoverable to each other through search);
2. Curate knowledge (for instance, semantic searches and computer vision that identify the most relevant content, and cluster it for people to process it more easily);
3. Collaborate across those entities now enhanced by the new knowledge (for instance, natural language processing that automatically translates content or machine learning that optimizes video and voice transmission);
4. …and Compute any other prediction (for instance, to determine which participant in the network is worth rewarding, or other machine learning algorithms to detect spam and fake inputs)

We don’t know the answer for sure, but work from MIT and others over the year hints at the possibilities: the creation of a networked, connected intelligence that could make our organizations smarter, and could turn organizations into intelligent systems — able to sense, create alternatives, act, and learn — that is adapt, over time.