How ‘Hybrid Intelligence’ can transform innovation management in modern organizations.
“Innovation distinguishes between a leader and a follower.” –Steve Jobs
All over the world, organizations are in desperate need of innovation. However, their teams are unable to deliver the level of innovation needed. This creates the ‘innovation gap’ in organizations. The major reasons for this gap are the human limitations. Hybrid Intelligence, in which human innovators work along with AI, can help to fill up this gap.
Why organizations need innovation
Life of an organization toggles between two modes: routine and problem-solving. It is the routine operations that earn the revenue. But problem solving makes routine operations possible. While problem solving exists in every area of work, problems in the following areas can have major impact on an organization:
● New product offering
● Operational efficiency
● Marketing and sales
● Finance
● Human resources
Innovation is closely related to problem solving. A solution is called innovative if:
● The problem solved is important
● Either there is no existing solution or existing solutions are unsatisfactory
● The solution manages the constraints of the organization
● The solution is well implemented
This makes innovation such an intrinsic part of an organization’s life.
The innovation gap
Innovation management (IM) teams are responsible for making sure that there is adequate supply of innovation in an organization. However, they face a number of challenges in delivering the required innovation. The major challenges are:
● Lack of suitable people: current innovation processes are dependent on people. To be able to innovate, individuals need a mix of skills that includes not only creativity but also adequate knowledge of the subject. Such individuals are not easy to find.
● Implementation difficulty: unless the innovative idea is implemented well, it is of no use. However, successful implementation requires problem solving to a detailed level. Most innovation teams find it hard to provide such a level of details. Thus many good ideas remain stuck in incomplete implementation.
● Right solution to the wrong problem: while the solution is quite creative and well implemented, the problem may not be an important one to solve. This poses the ROI problem which leaves the innovation unused.
Apart from the fact that it is not easy to find people suitable for innovation, there are some inherent human limitations that come in the way. Here are some of the major ones:
● Information processing limitations: managing the innovation life cycle successfully requires enormous amounts of information processing. For example, selecting the right problem to solve will involve sifting through heaps of feedback, complaints, suggestions, reviews and so on. Humans tend to rely on their intuition rather than processing all this information.
● Search limitations: Each phase in the innovation life cycle requires ‘search’ capabilities. For example, given a problem, the innovator has to first come up with a lot of solution candidates. Some of these candidate solutions will be well known while others will be alien. Humans find it difficult to achieve this process, called the ‘space search’. While most people usually identify the solutions that are near to their experience, only a few creative ones can go beyond the obvious and find far away and unrelated solutions.
● Expert bias: an organization has many experts in the domain who run the company. However, experts find it hard to provide novel solutions as they are deeply affected by what they know. This is a variation of the ‘search limitations’.
While the above discussion offers some explanation why the IM teams fall short of supplying the innovation demanded by the organization, we should not forget that problem solving and innovation happens in all teams, the IM team is just a facilitator.
The shortfall of innovation gives rise to the innovation gap in organization. This gap adversely impacts the organization’s ability to survive and prosper in volatile business environments.
“The enterprise that does not innovate ages and declines. And in a period of rapid change such as the present – the decline will be fast.”
– Peter Drucker, Author, management consultant (1909-2005)
How generative AI can help to bridge the gap
Generative AI is a branch of Artificial Intelligence. It covers certain types of machine learning (ML) models such as Large Language Models (LLMs), Vision Transformers, Speech-to-text models and so on. What is important for us is the fact that the models are trained on enormous amounts of data. This makes the models capable of producing meaningful and intelligent content such as text or images. Some common uses of generative AI models are answering questions, comprehension, translation, writing prose, summarization, creating images and so on.
Gen AI models are very good in the two areas in which humans face limitations: information processing and space search. This makes them perfect companions of humans for the purpose of innovation management. In fact, humans and AI working together can be the ideal strategy for filling the innovation gap. This way of working is called Hybrid Intelligence (HI).
Here are two examples of how GenAI can help to overcome human limitations.
Information processing in problem identification
The first problem in innovation is identifying the suitable problem to solve. Many organizations use the crowdsourcing method to generate ideas, in which stakeholders contribute suggestions and ideas. However, this gives rise to a vast pool of ideas which humans find difficult to sort and select. GenAI models can help in this task by doing the following:
● Create a clean list - remove duplicates, group the items under categories.
● Score the items - assign a score for each item according to the importance.
● Combine - merge related ideas to shorten the list.
Note that scoring an item requires specific knowledge of the organization. This knowledge can be added to the model using various methods such as finetuning and RAG.
Searching the solution space
A problem can have many possible solutions, some are proximate and many are not immediately obvious. Here are some possibilities:
● Already tried solutions
● Solutions from the same field
● Solutions from other fields
As an example, take the problem of reaching more customers in a software company. The innovators can consider solutions implemented by other software companies, but they can also see what other B2B companies are doing. Distant solutions can include the methods used by filmmakers to lure audiences.
Previously we mentioned that humans are good at coming up with familiar solutions from their experience. But GenAI can help to list solutions found in various sources such as case studies, white papers and books. They can wander far away and find solutions from unrelated domains. Models can also combine solutions to create really novel candidates. This helps to expand the solution space much wider than what human innovators can achieve.
Democratizing innovation
If human expertise is combined with the power of GenAI as the above examples illustrate, it will empower every person in the organization to innovate. This brings about a transformation in innovation management in the form of democratization of innovation. IM teams can garner far more resources than previously available.
Implementing GenAI for innovation
Organizations are laying down policies and practices for implementing AI. The major decisions revolve around:
● Local models or APIs: While Model APIs (such as OpenAI) make AI deployment easy and inexpensive, deploying models locally keeps the organization data secure.
● Build or buy: many AI teams are building their own software to power AI applications such as the ones described in the previous sections. Other teams prefer to use existing teams or involve a partner for the development.
● Managing ethical issues: AI models are fraught with risks such as bias and hallucinations. The leadership and AI teams need to provide practical solutions to these issues.
● Managing repercussions on the future of jobs: The human resources need to be trained appropriately to make HI possible. There is a fear about AI in the workforce that has to be handled properly.
Summary
Hybrid Intelligence takes shape when humans and AI work together. HI can be effectively used to overcome the innovation gap in organizations. IM teams can take advantage of the new GenAI capabilities (with right implementation) to bring about democratization of innovation in their organizations.
For further discussion on the same, please reach out to Devesh Rajadhyax on LinkedIn.
Want to be a part of CXO India? Reach out to us at info@cxo-india.com!
