I joined TechCrunch a year and a half ago to cover the emerging machine intelligence ecosystem. In that time I’ve been incredibly fortunate to have been able to spend my days highlighting brilliant researchers and founders building the next generation of AI-powered products and services — a natural extension of my studies in cognitive science.
One of my main goals when writing was to preserve my own excitement when I uncovered something new about machine learning. On more than one occasion, half hour interviews with founders turned into three hour white boarding sessions. Unfortunately, no matter how emotionally invested I was in the work I was reporting, I couldn’t always get past the caution and restraint founders maintain when dealing with journalists.
Fortunately I was surrounded by a team of people full of smart opinions and commentary on the tech world who demonstrated to me just how much one can accomplish on the back of words. I’m grateful to my amazing colleagues at TechCrunch who helped me navigate this world and taught me how to remain skeptical about technology without becoming cynical.
But even so, it’s easy to get antsy when you have ringside seats to one of the greatest technological trends of the last century but can’t get your hands dirty building. With this in mind, I’ve decided to hit the gym and get in the ring at Basis Set Ventures, a brand new $136 million venture capital fund targeting early-stage machine intelligence startups.
Moving forward I’ll be spending time around many of the same types of startups that I worked with while at TechCrunch. This includes the expected, those making use of recent advances in neural networks and deep learning to solve big problems with computer vision and natural language processing. But it also includes less obvious implementations of wonky, obscure and sometimes even retro machine learning models that help enterprises take advantage of AI without armies of scarce data scientists and copious amounts of data.
I’m incredibly fortunate to be joining Lan Xuezhao, founding partner of BSV, at a time when the AI ecosystem is still fresh enough to accommodate a new opinion or two.
I first met Lan in August and was immediately impressed with her lack of ego, original thinking and practical approach to problem solving. I share her vision and excitement for verticalized applications of artificial intelligence that prioritize product and commercial viability.
While the nearly 500 stories I wrote while at TechCrunch were the most visible component of my work, writing only consumed a fraction of my time. A majority of the words I’ve put out were written in the early morning, often as early as 5am. What most people consider to be the work day was mostly spent taking meetings, performing diligence and brushing up on AI concepts.
Between my studies, public service and time at TechCrunch, I’ve spent a lot of time trying to figure out why people do what they do. I still come up empty handed pretty often but I’ve come to appreciate the important roles that incentive structures and human behavior play.
The founders that get me most excited start here when articulating their vision. Every piece of software released influences people whether we like it or not. We owe it to users to be deliberate rather than reactive when bringing disruptive tools to market. Creators should strive to simultaneously nudge individuals, groups and industries to achieve greater outcomes.
The first story I authored for TechCrunch was written from my dorm room. I talked about the ability for technology to influence human behavior citing startups like OPower and Timeful. It’s 2017 and Dan Ariely, one of the key architects of the behavioral science movement, is supporting startups at Google Launchpad’s AI Studio. That says something.
Technology narratives shaped by publications like TechCrunch are incredibly important in influencing public opinion and ultimately dictating the direction of the industry. The same is true for venture capital. I strongly believe venture capital to be another critical point of leverage over innovation.
I’m excited to help build Basis Set Ventures while continuing to examine key drivers in the innovation economy. Helping to shape the commercialization of artificial intelligence today will impact the workplaces, factories, global supply chains and labor norms of tomorrow.
AI on its own may not be a business model, but it is a resource intensive enabling technology. From a venture capital perspective, AI startups are unique entities with specialized needs that can’t be ignored. I could not be more enthusiastic to start meeting founders who are savvy product specialists with strong, succinct and practical plans to capture massive markets.
It’s day one. My inbox is clean. It’s ok if you’re only machine learning aspirational at this point. I only ask for honesty. Let’s do this.