Applied AI & AI in business /

Increasing the deep learning footprint: From niche to mass adoption

November 19 2018/4 min read
  • Rob Dalgety
    Rob DalgetyProduct Marketing Director

AI has enormous potential to impact the lives and well-being of people worldwide and reshape how organizations run their operations and interact with the outside world. But right now, AI is mainly in the domain of tech pioneers and large businesses. To reach its potential and make a real impact on business and society, AI must be made affordable and usable for any organization.

Making AI, and especially deep learning, usable, accessible and affordable for all is our goal at Peltarion. The journey to get there requires a new type of solution: an end-to-end operational AI platform that will abstract up from some of the complexity currently bogging down deep learning efforts and facilitate more AI projects, enabling a wider range of users to be applied to many AI/deep learning projects where deep learning methods would add value.

Increasing the footprint: problem types, users and organizations

The tech pioneers, big companies with deep pockets and academic institutions will always continue to push the boundaries of what’s possible and achievable with AI. But operational AI technology will enable deep learning to be integrated into real industry and humanitarian efforts right away. Current AI technology cannot solve many of the existing challenges to effect real, significant and positive outcomes. Operational AI platforms, on the other hand, are targeted at the vast array of problems where deep learning could be applied and deliver value.

Because operational AI covers every key part of the deep learning project workflow and abstracts up from the underlying complexity, it creates a wider footprint of potential users where an expanded set of people within an organization’s technology team can apply deep learning methods. This enables junior data scientists and even those who aren’t deep learning experts to nonetheless use deep learning methods to solve problems. (Operational AI would still give the top echelon of data scientists the ability to solve problems faster and achieve quicker modeling and evaluation, too.)

Operational AI platforms are targeted at the many problems where deep learning could be applied and deliver value

Operational AI also creates a wider footprint of organizations. Because it enables additional types of users to run deep learning projects, many more types of organizations can now start to apply this technology. It helps overcome the acute skills gap in AI and in deep learning. At the same time, the platform allows for both an increased and more repeatable flow of AI projects to be run and then actually implemented. What has, until now, been only the domain of big tech pioneers and academia becomes accessible to organizations of all types: mid-tier enterprises, NGOs, governmental bodies, educational organizations and startups. The real value is not just getting one AI project implemented but organizations embedding AI and impacting how they interact externally and run their operations.

More users, more organizations, more problems solved by deep learning — that’s a much wider footprint for AI and the start of our journey to make AI usable and affordable for all.

Want to learn more about how to move from research to reality for deep learning and AI? Read the whitepaper The case for Operational AI.

  • Rob Dalgety

    Rob Dalgety

    Product Marketing Director

    Rob Dalgety leads Product Marketing at Peltarion. He has extensive experience in commercializing and positioning software into enterprises and other organizations in areas including mobility, big data and analytics, collaboration and digital.

02/ More on Business & AI