IRVINE, Calif. — “Becoming a software company is really hard,” said Vijay Gurbaxani, founding director of the Center for Digital Transformation at the University of California, Irvine.

However, failing to become a software company could be detrimental to the outlook of any business. It’s important for enterprises to focus on making software work for their benefit, not against it, he said in the opening remarks at the Road to Reinvention conference. The wave of improvements in machine learning during the last few years have been amazing, but it’s also opening a new era of relations between humans and software that cannot be ignored, Gurbaxani explained.

“New knowledge is no longer going to be restricted to humans conceiving of theories,” he said. It will be driven by machines and “it’s only a matter of time when machines can outthink humans. What do we do in a world where know-how is created by machines?”

Software is overdue for a high-level review, Gurbaxani argued. While most people work in companies where software is increasingly critical, relatively few are thinking about the impacts of software in a holistic way, he said. “Society has increasingly gone on autopilot.”

The books people read, movies they watch, people they date, and restaurants frequented are all increasingly determined by code and algorithms, but that factor is rarely scrutinized, Gurbaxani explained. “We’re also using [code] to determine what a jail sentence should be for a criminal” and  “we don’t really know what is in these systems.”

With software reaching such an increasingly pivotal role in society and business, the impact of software and how it calculates outcomes needs to be analyzed much further, he said. “These are things that need to have light shined on them.”

Data is still relatively disconnected and businesses are struggling to make decisions without knowing all the factors at play, Gurbaxani explained. “If you want to build a system that is really useful to us, you have to link all this data.”

“We’re not going to get there if we don’t have trust in the platforms and we don’t have trust in the data,” he said. “We must rethink the system design” and “we can design a system where it works for everybody.”

Robots should be about serving humans, not the other way around, Gurbaxani said. “Those are the kinds of conversations we should be having every day... We must put people first.”

While software could lead to unimaginable negative consequences, it has also proven itself capable of doing work that humans don’t engage in today, he said. “Rather than focus on job loss… I would much rather focus on what is possible.”

Enterprises and consumers should be striving to solve problems that humans are incapable of solving, he explained. “Let’s build a world that we want to live in.”