Creativity is a key ingredient in idea generation. It can help you come up with solutions to problems you might not have considered before, and it’s essential in brainstorming sessions.
AI can automate this process, and it also supports innovation by enabling teams to explore large problem and solution spaces with greater speed and scale. However, the challenges of deploying AI in this way are many and complex.
Machine Learning
Machine learning is the part of artificial intelligence that deals with enabling computers to learn without being programmed explicitly (see Figure 1 above). This type of AI involves machines studying examples and using those examples to predict how to perform new tasks. It is a core technology behind familiar systems such as E-commerce websites’ recommendation systems, Netflix and Spotify’s movie/music recommendations, and Amazon’s Audible’s personalization algorithm.
Another type of machine learning is natural language processing, which enables computers to recognize and understand the words that humans speak or write. This enables technologies like chatbots and digital assistants such as Siri and Alexa. It also enables text analysis, such as the highlighting of certain words or phrases in a document or a webpage, and translation between languages. This type of machine learning is often based on neural networks, which are modeled after the human brain in which thousands or millions of processing nodes are interconnected and organized into layers.
The ability of machine learning to process large amounts of data at a faster rate is also helping companies with data analytics, allowing them to find anomalies and significant patterns in business information, which can lead to opportunities for innovation. Examples of this type of machine learning include the use of pattern recognition by Celonis, which finds organizational processes that could benefit from robotic process automation, and Outlier, which identifies trends in data to identify outliers in customer behavior.
Despite the potential of these emerging technologies, it is important for innovation managers to be aware of their limitations. They should avoid looking at a particular piece of technology as a solution in search of a problem, and instead begin with a focus on identifying a business problem or need that can be addressed by an emerging technology.
This is especially critical for ensuring that the underlying technology can achieve the desired result. For example, it would be unacceptable if an AI system only performed to 95% of human accuracy—that level might be okay for a movie recommendation engine, but it wouldn’t be enough for a self-driving car or a program designed to detect serious flaws in machinery.
Natural Language Processing
Natural language processing aims to narrow the communication gap between computers and humans. This involves the use of algorithms to transform human language into something that a computer can understand. This process includes tasks such as identifying words and syllables, analyzing context, removing stop words (words that add little or nothing to the meaning of a sentence), and understanding word ambiguity and grammatical structure.
As a technology, NLP has come of age over the past ten years. It is used in products such as Siri, Alexa, and Google’s voice search. It also supports many other applications, including customer service, medical research, risk management, insurance, and contextual advertising.
In addition, NLP is an important component of machine learning. It helps provide the clean, curated data that is needed for advanced predictive models. This is especially important when using data from unstructured sources such as electronic health records, clinical trial records, or full text scientific literature.
NLP is a broad field that is incredibly complex, and its effectiveness relies on many factors. NLP is also a constantly evolving technology, and it is likely that the rules of language will change as we learn more about how people use language in different contexts. As such, NLP systems are subject to error and may become obsolete over time.
One of the biggest challenges with NLP is determining the meaning of words in their natural context. NLP tools need to go beyond dictionary definitions and syllable counts to understand subtleties of tone, inflection, and nuance in speech or writing. This is a challenging task that requires a deep understanding of the human mind and the ability to interpret emotions and motivations, such as sarcasm.
Another challenge is ensuring that NLP is fair and equitable in its decision making. For example, an algorithm can make biased decisions if it is tasked with reviewing legal documents or interviewing applicants for a job. Natural Language Processing in Brainstorming is an important challenge because NLP systems can affect the lives of millions of people and impact their careers, health, or quality of life. Startups such as Verneek are working to create software that can detect bias in the decisions made by NLP programs so they can be corrected.
Artificial Intelligence
Artificial intelligence encompasses a broad set of capabilities that mimic human thinking, including reasoning, problem solving, planning, learning, acting, reacting and understanding and generating language. The newest AI systems have evolved beyond the basic rules-based machines that first emerged in the 1950s, when they became known as “expert systems.” These sophisticated programs can recognize patterns and trends in large data sets to make informed decisions. They also learn from past experience and adapt to new situations. Examples of this are GPS apps that give updated turn-by-turn directions in almost any part of the world and IBM Watson, which defeated Grand Masters of chess and champions on Jeopardy!
Currently, AI technology is in use at the highest level of business operations. Many organizations have adopted AI as a strategic imperative to achieve more in less time, create new revenue opportunities and boost customer loyalty. However, implementing and managing AI at scale is complex and requires the right tools, processes and management strategies to succeed.
One of the most promising applications for artificial intelligence is in creative industries. Already, AI is helping write pop ballads and has simulated the styles of painters and musicians. It is possible that AI could someday be a true creative partner and even create solo works of art. Experts disagree, however, on whether this will ever happen.
In the meantime, the most practical use of AI for creativity is to add it to existing products and services. This can improve performance and reduce costs, and it’s common for businesses to integrate artificial intelligence into their websites or chat portals.
For example, an Italian insurer developed a cognitive help desk for its employees that uses AI to find solutions to problems. The system uses natural language processing to understand requests, search for documents and previous answers, and a decision-making function to automatically route questions to the appropriate person.
AI is transforming knowledge work by automating repetitive tasks and enabling rapid, high-volume discovery through data. It is enabling faster, more reliable, and repeatable processes that free up humans to focus on higher-value work. This may eventually reduce the need for highly-skilled labor and replace routine activities in many sectors, but it is unlikely to fully replace human management, as suggested by Cyert and March.
Big Data
The ability to collect and process data at an unprecedented rate is creating a wealth of information for organizations. Many of these data sets are too large to be handled by traditional data-processing application software, but they contain valuable insights for companies looking to improve operations or create new products and services. This data can help them better understand customer behavior, predict market trends and other important information that may be otherwise difficult to find.
Big Data is a voluminous set of structured, unstructured and semi-structured data that’s challenging to manage with traditional tools and requires additional infrastructure to govern, analyze and convert into insights. This data comes from a variety of sources, including internal systems like application and server logs, social media posts, external APIs, clickstreams, sensors on physical devices, geographic information, scientific research and images. This data is typically streaming and produced on a continual basis, meaning it needs to be quickly analyzed and ingested in order to be effective.
As a result, Big Data requires faster processing and scalability than other types of computing. This is why cloud computing, with its elastic scalability and built-in features for data management, has opened the door for companies to use Big Data at scale. In addition, the emergence of graph databases has helped Big Data solutions become more accessible and easy to use for developers.
Organizations that successfully leverage Big Data can uncover insights they can use to drive innovation and competitive advantage. They can make better decisions that improve the effectiveness of their operations, create personalized marketing campaigns and provide a more individualized customer experience. Big Data can also help them identify at-risk patients so they can receive treatment sooner, prevent crime by predicting where a criminal will strike and improve supply chain efficiency by tracking inventory to avoid outages.
Big Data can transform business processes, accelerate the time to market for new products and services and increase revenue and profits. It can also help organizations improve human resources, technology and marketing initiatives by providing more accurate forecasts about consumer demand and product performance. It also allows them to identify opportunities for cost savings by reducing unnecessary expenditures.