Key Takeaways

• AI pioneer Tom Kehler laid the brickwork for artificial intelligence way back in the 1980s. 

• Kehler says the third wave of AI will be human-empowered, with highly intuitive tools capable of abstract reasoning. 

• Read how  Kehler applied the rules of neuroscience to analyze startup success, driving investing in companies run by women and minorities. 

  
  

In the 1980s, at the height of the personal home computer frenzy, he laid the brickwork for the next leap in technological progress: artificial intelligence. Four decades later, AI pioneer Tom Kehler is predicting another breakthrough in a field he has dedicated his life to. 

A preacher's kid, Tom was interested in studying linguistics with the Wycliffe Bible. But he said that God kept closing that door on him. After earning a Ph.D., Kehler chose natural language computing instead. While he continued feeding his passions in the academic world, he got a call from Silicon Valley. A professor called Ed Feigenbaum recruited him to be part of a new startup spun out of Stanford focused on applying AI to genetic engineering. Hereon, there was no looking back. Tom Kehler's career would witness insurmountable heights in the decades that followed.

In the mid-1990s, when the internet was still in its infancy, Kehler became the CEO of Connect, an e-commerce platform. He went on to patent algorithms for giants NBC, General Motors, LEGO, and Procter & Gamble. He recalls becoming ‘obsessed’ with using the internet to understand humanity's ‘collective intelligence.' In 2015, he co-founded CrowdSmart AI, a SaaS collaboration platform that used AI to predict startup funding success.

In this conversation with Horizon Search, Kehler talks about the fast-approaching wave of intuitive artificial intelligence, the free energy model and its application in evaluating startups, and the future of his company CrowdSmart AI.

The Next Wave: Human-Empowered Artificial Intelligence

“Imagine if the next breakthrough in artificial intelligence came from the root of intelligence itself: the human brain.” — Tom Kehler

While some experts have predicted that artificial intelligence will amplify human effectiveness, others have been vocal about the more damning aspects of this technology. In question is human autonomy, agency, and capability when placed in the lap of machine learning's unregulated power. But Kehler, a veteran riding on the wave of AI innovation, wants you to reimagine its role in human society. 

His views cease to reflect the dystopian future critics often forecast. An AI architecture, recognizing the function of the human brain, is our future. Humanity is capable of building AI respectful of the environment humans constantly interact with. “We should empower it [AI] with human knowledge… so that it operates in coordination with evolutionary and biological principles,” he says. What Kehler is referring to is the third wave of artificial intelligence. Staff writer for Futurism, Roey Tzezana illustrates the defining factors of each of the three waves of AI. These include: 

AI 1.0 - Handcrafted Knowledge: In the first wave of AI, experts devised algorithms and software by the knowledge they possessed. First-wave AI systems were usually based on clear and logical rules.

Examples: Creation of chess-playing computers and delivery optimization software.

AI 2.0 - Statistical Learning: In second-wave AI systems, engineers and programmers don’t teach precise rules for the systems to follow. Instead, there’s the development of statistical models for certain types of problems. Then, these models are ‘trained’ on various samples for precision and efficiency. The poster boy of the second-wave systems is the concept of artificial neural networks.

Examples: Face recognition, aerial drones, identification of objects in pictures.

AI 3.0 - Contextual Adaptation: This third wave, cited as human-empowered AI by Kehler, involves AI systems constructing models that explain how the world works. In other words, these AI models are armed with discovering logical rules that shape their decision-making process. They'll even be able to program and potentially develop abstract thinking. 

Example: The German Defense Ministry, last year, invested in the creation of an AI-based military “metaverse” called GhostPlay. It utilizes advanced third-wave algorithms to facilitate more “human-like” decision-making for simulated units, setting itself apart from traditional second-wave algorithms. 

Blast from the past: AI pioneers John McCarthy and Ed Feigenbaum discuss the emergence of AI's first wave. Tom demonstrates the KEE system, a precursor to modern AI technologies in 1984.

Science & Investment Collide: On Startup Survivability

Karl Friston is the world’s most cited neuroscientist with a h-index (a metric used to measure the impact of a researcher’s publications) nearly twice the size of Albert Einstein. The neuroscientist has identified the organizing principle of life and all intelligence. Enter Friston’s free energy model which says that all life, at every scale of organization–from single cells to the human brain, with its billions of neurons–is driven by the same universal imperative, which can be reduced to a single mathematical function. “To be alive is to act in ways that reduce the gulf between your expectations and your sensory inputs,” he says. 

The model has numerous implications. Some want to use the free energy principle to unify theories of the mind, provide a new foundation for Biology, and explain life as we know it. Others hope the free energy principle will ground Psychiatry in a functional understanding of the brain. Most importantly, Friston’s ideas serve as a breakthrough in the study of artificial intelligence. 

When applied to business, the free energy model can predict the survivability of startups. Tom has helped 100 different companies in this arena. He enabled them to deduce whether profitability is achievable through their investment plans.“We were able to show that about 85% of a group [in a startup] can predict its survival rate,” he claims. Kehler's intervention was good news for most startups. “70% of those companies we invested in are still alive through Covid,” he said.

The AI pioneer went on to transform how investment decisions were made in the startup ecosystem through anonymous evaluation, data-driven analysis, and Bayesian sampling. He ensured impartiality by choosing not to reveal the identities of startup founders during the evaluation process. Kehler’s team then leveraged the Bayesian sampling to objectively assess startups based on metrics that matter: market trends, past performance, and team capabilities. Eventually, this led to a pathbreaking result. His team ended up investing in 42% of the companies that were run either by women or minorities, something quite unheard of with venture capitalists.

AI For Common Good: Future of Crowdsmart 

Kehler wants to merge his brainchild CrowdSmart AI with an entity called Common Good AI. “The idea is to open up the architecture so that other people can use the API,” he says. In the longer run, his dream is for CrowdSmart to play a role in preserving democracy “I want it to be the opposite of social media. That’s the dream,” he says. 

Despite being successful in taking two companies public and being deeply entrenched in the startup ecosystem, Kehler prefers people not to call him a businessman. “I’m not your business CEO type,” he says wryly. “The worst thing that happened to me is I got suckered into it [being CEO] in my ‘30s,” he continues. Kehler will always be a scientist and innovator first. “I have much more joy in creating technology rather than running companies,” he declares.