Many venture capitalists and technologists talk about artificial intelligence, big data and how everything from systems to objects are getting smarter. Devices talk to one another, and people are learning how to talk to devices. Google Home, Alexa and Siri are just a few examples of how spoken language can help automate our lives to get things done more quickly. Businesses are evolving, too. Objects gather information, software is developed to make decisions, and our virtual assistants make sure it all runs smoothly.
It’s inevitable that this technology also will have a profound impact on the way businesses are started and accelerated. For example, the $9 billion dollar company Stripe already has automated the very complicated process of starting a business. Its Atlas product incorporates a company in the state of Delaware, sets up a bank account with tax ID number, and establishes a Stripe profile so the new business can accept payments from customers.
In the future — as now — entrepreneurs still will make many final or otherwise crucial decisions. But consider, for a moment, all the little decisions that cloud your working and private lives. It could be something as simple as finding a good Italian restaurant near your home or office. Ask your phone, and you’ll get a list of all the choices. Why doesn’t it filter out the obviously bad options? Why can’t the software determine it needs show only the closest location of a franchise, for example? Or take into account differentiators such as price, distance and user ratings when two choices have nearly identical menus?
Context and value.
Humans want good choices, not 100 choices. And big data’s sheer wealth of information creates decision fatigue at the company level as well. Corporations collect every little detail, but the petabytes of data stored on hard drives often are cleared at the end of the year because business leaders don’t know how to interpet the information or what to do with the knowledge.
Current methods require customer-engagement surveys via email, phone calls or snail mail to gather enough of the right kind of information to generate moderately actionable results. These approaches are costly in terms of company time and resources, and the exercise itself can create customer fatigue. There aren’t enough IT people to support all this information, once the business collects it. A huge opportunity exists for software startups that unravel the information-overload problem for customers and companies alike.
Natural language processing.
Natural language processing (NLP) offers a promising solution. Software built with NLP allows computers to “read” and process language in a way that enables software to carry out research that previously required humans to conduct phone surveys, complete database entry and run database queries.
It’s no secret that corporate America often favors the younger employee with more computing experience over the mid-career worker who brings more field experience but finds it difficult to adapt to new technology. Voice-based interfaces such as those used by Siri and Alexa let employees step back from the need to develop a deep technical background. NLP technology empowers people to focus on the tasks that computers can’t duplicate. Instead, NLP allows computers to become “the IT person.” Software built with NLP is ideal for companies that value domain expertise and are shifting their cultures to align with that principle.
What humans do better than machines.
This is a win not only for those who struggle to keep up with changes in software but also for corporations as a whole. Managers can devote less time to teaching workers how to use computer programs and more time to educate employees on customer service, domain knowledge and proven sales techniques, to name a few.
When software does the heavy lifting to compile data and author reports, human team members are free from the sort of drudge work lampooned in “Office Space.” Users will add commentary that the software can’t possibly know, providing real insights and analysis. For example: “We expect advances in such and such to reduce this expense next quarter.”
As more devices and systems communicate and senors become more widely available, so will the data itself. Then, programming begins to take the form of simple “if” and “when” statements: “If it starts to rain, close the windows. When it stops raining, open the windows if the outside temperature is between 68 and 75 degrees.”
Effects on education and society.
This degree of NLP automation is leading us to a society in which college graduates use pseudo code: English language to embed logic statements. In the future, job titles will change because the members of the work force will have learned how to automate their own jobs. Employees will redefine work and occupy a number of roles simply because they can be more productive. By definition, then, this new generation of workers will be an entrepreneurial one.
It’s no stretch to imagine schools that teach first graders the basics of logic and coding — not in the traditional format, but through procedures students enter via voice commands and written language. By the end of second grade, rote memorization could be far less of a priority. After all, most answers are just an ask away on the internet, and the new science and social-studies curriculum looks a lot like information retrieval and computer programming.
Computers that ‘think’ like people.
In my own work, I enjoy creating computer programs that are based not in math but in pattern recognition. It’s how human intuition works. People aren’t computers, and computers aren’t people. Still, people have tried for years to make computers learn human language by manipulating the tasks computers handle exceptionally well.
As an entrepreneur, I’m passionate about building technology that flips the approach: Let’s teach computers to understand language using techniques that mimic neurology and psychology. The technology is set to unlock all the data on the internet, in your computer or waiting within your corporate network.
Let’s make your children’s homework so easy that we abandon memorization drills and instead focus on developing their critical-thinking skills. The future of work centers on elevating what people do well, not dwelling on all the things computers can’t yet deconstruct