Fixing Innovation

Modern fields of science and engineering are driven by the hypothesis method. This method involves careful observation and the application of rigorous skepticism about what is observed. It also involves formulating hypotheses based on such observations, experimental and measurement-based testing of deductions drawn from the hypotheses, and refinement, or elimination, of the hypotheses based on the experimental findings. [^1]

In modern research & development (R&D) projects, researchers follow the Parsimony Principle, also known as the Rule of Simplicity. The Parsimony Principle states that the hypothesis should be as simple as possible but not simpler. As a result, science and engineering are driven by a mode of inquiry that focuses on identifying a relatively small number of causal drivers of underlying phenomena built upon an underlying theory. Modern companies and research academia, consequently, adopt this mindset when developing and launching projects focused on innovation.

Traditional innovation is broken.

  1. Modern innovation requires an engineering talent pool. The labor costs quickly add up as companies have to write the algorithms and maintain it over time.

  2. Automation-oriented innovation requires if-then application development. As business logic is written in code, technical debt quickly adds to the overhead of managing modern software projects.

  3. Software-driven innovation is unable to deal with the changing volume, velocity, and variety of data. Companies are able to store data, but they are not making insights and predictions on that data without involving algorithms or specific technologies.

  4. While R&D investments are increasing, R&D returns are decreasing. [^2]

  5. The vast majority of innovation projects fail. [^3]

To catalyze innovation, companies have invested billions in internal venture capital, incubators, accelerators, and field trips to Silicon Valley. Yet according to a McKinsey survey, 94% of executives are dissatisfied with their firms’ innovation performance. [^4] Across industries, one survey after another has found the same thing: Businesses just aren’t getting the impact they want, despite all their spending. Fortunately, it’s possible to hack innovation.

The arrival of deep learning technologies introduces a new research paradigm - using deep learning as a general-purpose method of innovation. Deep learning offers an alternative paradigm based on the ability to predict complex phenomena. Deep learning models abstract away from writing algorithms and instead focus on gathering data. Deep learning technologies do not require domain experts. Instead, deep learning or neural networks force companies to focus on data gathering and data management. In other words, deep learning is an innovation in the method of innovation.

What are the advantages of using deep learning technologies for innovation management and maintenance?

Deep learning allows Fortune 500 companies and global enterprises to:

  1. Focus on fixed-cost investments in AI instead of highly skilled labor

  2. Improve performance in existing research projects

  3. Reduce the level of technical knowledge required

  4. Derive new insights

  5. Predict complex multi-causal phenomena [^5]

Deep learning-driven innovation is the only way companies can survive in the modern competitive landscape.

Interested to learn more about AI-driven innovation?

Schedule a call with our team!


[^1]: Wikipedia Contributors (2020). Scientific method. [online] Wikipedia. Available at: [Accessed 2 Mar. 2020]. [^2]: Capgemini (2015). The Innovation Game: Why & How Businesses are Investing in Innovation…. [online] Available at: [Accessed 2 Mar. 2020]. [^3]: Capgemini Worldwide. (2015). Capgemini Consulting and Altimeter global report reveals leading businesses continue to struggle with innovation, with traditional R&D model ‘broken.’ [online] Available at: [Accessed 2 Mar. 2020]. [^4]: McKinsey & Company. (2020). Leadership and innovation. [online] Available at: [Accessed 2 Mar. 2020]. [^5]: Cockburn, Iain M, Henderson, R. and Stern, S. (2018). The Impact of Artificial Intelligence on Innovation: An Exploratory Analysis. NBER, [online] pp.115–146. Available at: [Accessed 2 Mar. 2020].