Innovation: Can Algorithms Innovate?

(Area: Innovation, Technology, Collaboration)

(Reading Material "Innovation Advanced Course: Innovation in the New Age", Hugo Céspedes A.)

This is a topic that I wanted to touch on for quite some time, for which I will address three topics such as: Data and Algorithms, Innovation and Artificial Intelligence.

Artificial Intelligence "has generated creative works of art (pictures)"; "helping users through technological gadgets, in the role of virtual assistants, such as Google Home or Amazon Echo in homes around the world"; "on smartphones continuously, for example when selecting virtual assistance via voice request, or when selecting portrait mode in the camera of the same smartphone, among many other actions"; "by continuously analyzing data in companies to learn about our habits, through the use of Big Data and Artificial Intelligence"; "in the field of medical applications, where machines working with artificial intelligence help doctors where the doctor's clinical eye cannot do it, and in this way we can enjoy defibrillators, surgical machines, diagnostic machines where AI helps to offer better results"; "in the logistics sector, where AI is used to optimize routes, with better alternatives for travel (based on comparisons of geographical data, of the environment in real time)"; among many others.

The Data.

As Diego May argues, today there is a "new currency" that is helping the largest Exponential Organizations (ExO) in the world to grow disproportionately. This currency is used by the world's leading technology companies, not only to achieve success, but to build lasting dynasties and become legendary. Facebook, Google and Netflix are just a few of the many companies that take advantage of this "currency" (which does not refer to cryptocurrencies). This new currency is "the Data" (which today resembles the old observation used to design and Innovate). Today, data has become the new oil. It is the most valuable resource in the world. And while technologies like smartphones and the internet have made data abundant and ubiquitous, those who succeed will be the ones who know how to take advantage of the data they have access to.

Data presents an invaluable opportunity for organizations to innovate, but only if they know what to do with it. Wharton Professor of Operations, Information and Decisions, Lynn Wu discusses how different organizational structures influence the use of data analytics to drive innovation. Her article, "Data Analytics Support Decentralized Innovation", touches on this topic

This is where "Algorithms", one of the 11 attributes of ExOs, come into the picture to help organizations make sense of vast amounts of data and apply it to scale operations and profits (and clearly, innovate to improve the quality of life of customers and users). The best part is, "You don't have to be a technical company to take advantage of Algorithms, as it's within the reach of any executive or entrepreneur". This is where our topic arises: Can Algorithms Innovate?

Innovation Today.

In a previous post, I had already developed the subject of "Pre-Pandemic Innovation ". First, I will begin with the concept of Innovation (on which I will base myself), although I will not deepen into the subject (If you want to know more, I will leave you with some material here: "Innovation "; "Social Innovation: Does it arise only from Social Entrepreneurship? "; "Open Innovation "; "Ethnography and Innovation "; "Creativity and Innovation: How to Break Paradigms "; "How to Be More Creative and Not Deviate from the Path Towards Innovation "; "How to Measure Innovation and Why Measure It: Part One "; "How to Measure Innovation and Why to Measure It Part Two: The 7 Deadly Sins "; "Innovation Cash Curve "; ..)

I will only say that the definition that I will use to refer to Innovation in this post is related to the fact that "Innovation refers to the ability to Create Value for people -customers, users, citizens, beneficiaries,...-, to organizations -profit and non-profit-, for society in general, considering the minimization of negative externalities based on our innovative design, considering a positive impact at an individual and social level, improving the quality of life of people and society in a substantial way, as well as the environment". I do not include the concept of generating monetary benefits, since I always counter question clients, students, mentees: What would happen if tomorrow the money economy collapses because users decide from one day to the next not to use more money?", it disappears innovation" because its definition in monetary terms no longer exists? Nor do I like to define it in terms of Incremental Innovation, or Disruptive Innovation, or Creation of New Solutions, ... only, since they leave out a whole spectrum of Innovation that happens. That is why I like to define it from the point of view of its impact from the level of user, environment, society, environment.

Analytics and Innovation.

Lynn Wu speaks in the most general way about the ways that Data Analytics can help drive Innovation. "There are so many good examples. There are a lot of analytics technologies, especially driven by recent advances in Machine Learning and the sheer amount of Digitized Data".

It has already happened that "an analytics-driven machine was able to beat the best human player in the world at the game of Go". We also have driverless cars that depend on the vast amount of digitized images that drastically improve vision recognition systems. We've seen IBM's Watson sift through reams of research literature in digital registries, and find six new cancer suppressants in two months. Investigators would have taken years to find it. Even in the realms of art and music, for the first time, Artificial Intelligence -AI- analysis is creating art that people are willing to listen to, Wu says. So we see a lot of Data Driven Analytics creating a lot of really cool innovations around us.

In terms of how companies can use data analytics to drive innovation, what was the core question or problem you were trying to address with this research?

If you look at the innovation statistics, economists have documented that we've been spending more and more money per capita on research, but we're actually experiencing a decline in the rate of innovation that we're generating. "We're spending more money, but we're getting a lot less money back", Wu says. That seems like a paradox. We've seen lots of great innovations based on data, but we don't see innovation statistics.

"Analytics is really great at finding these hidden links or patterns that we can't easily see when pulling a ton of data".

That makes you think of a famous quote from Nobel laureate Bob Solow from three decades ago: He said "we see computers everywhere, but not in productivity statistics". If you replace computers with analytics and you replace productivity with innovation, we have the same paradox. What I'm trying to do in my research, Wu argues, is to see if we can use the same set of frameworks to explain the paradox of analytics innovation that we're seeing today.

In Wu's article, she notes that "there are many factors in an organization that could affect Innovation", but he focuses specifically on two ways companies handle Innovation: Decentralized and Centralized. What's the difference? While there are tons of factors that could mediate the relationship, Wu deliberately chose Decentralized and Centralized Innovation, because a lot of work has been done on the questions What is the advantage of decentralization and what is the advantage of centralization? What she and her colleagues define as "Decentralized Innovation Structures" are based on their collaborative networks. They can think of inventors creating the same patent and working together, so "there is a link between them". When they describe a Decentralized Innovation network, it's really about how concentrated the innovators are (when they collaborate). You can see a large group with many inventors working together, or you can see a "Very Decentralized or Dispersed Structure", the opposite of a Concentrated one, where there are many small groups of individuals and they are loosely connected to each other.

The reason why Wu chose these two structures is that "neither is absolutely better than the other". We have seen in many industries that highly innovative companies have both types of structure. If you look at Apple, it's very much a concentrated, centralized cluster with a small group of people responsible for the vast majority of innovation. But if you look at Google, you see a small group of clusters and they are loosely connected. They are also very productive in terms of innovation. You see that in the pharmaceutical industry as well: You see Sanofi and you see Roche .

Sanofi has "a much more decentralized or dispersed structure", and Roche has "a more concentrated structure". My question, was Do these structures play a role in how you use Analysis to Innovate?.

Wu says that “one structure is not better than the other”, but what did Wu find in his research, in terms of how these structures impact Innovation? That is getting into the key of what "Analytics can do for Innovation". What he found is that "analysis" can actually drive the creation of recombination, or combining a diverse set of existing technologies in a new way. Each individual technology already exists, but how do we somehow recombine them to create a new innovation? Or reuse something that we know solved a problem, but apply it to a different domain? Analytics is really great at finding these hidden links or patterns that we can't easily see when pulling a ton of data. That is really the key to driving Decentralized Innovation for several reasons.

"Analytics can really drive the creation of recombination, or combining a diverse set of existing technologies in a new way".

The Advantage of Decentralization is that, there are small groups working on a problem, so they really know what the problem is in that domain. They see what exactly they can do to solve that problem more closely than a centralized structure, which is larger but much more coordinated. A Decentralized Structure lacks coordination. They know very well what they are doing, but they do not know what others are doing. Centralized Structures know what everyone is doing, but they don't know the details of every single problem in the domain, unless they have the ability to analyze reams and reams of data to find hidden patterns. That is exactly the Disadvantage that Decentralized Structures have. In this sense, decentralization does not easily find other people's work. "Analytics" finds a way to select that and find a new combination for you, a new way to solve your problem that you may not have found easily without analysis. That ability, of course, can also help a Centralized Structure. A Centralized Structure already has this search and coordination mechanism incorporated; it just doesn't have a much higher marginal benefit than a Decentralized Structure would.

As Data Analytics Becomes More Pervasive, Will It Be Worth It For Organizations To Move To A More Decentralized Innovation Structure?

That's a great question. Wu thinks that depends on what his "Innovation goals" are. If you had a Decentralized Structure and you really want to do an Innovation that combines existing technologies in a new way, or reuses technologies applied to a different domain to solve a different problem, Analytics is great to help because you get both benefits. You delve into the problem domain and gain various insights from the outside.

However, Centralization is great for looking at bigger pictures and creating novel technologies that could act as a building block for future recombinations. They are fundamental technologies. That's hard to create with Big Data. Centralized Structures do not necessarily need to have big data to create that kind of technology.

Like Apple? Exactly. Much of it is human creativity or intuition, which is a bit difficult to digitize. If you were in that type of Innovation work, then having Analytics or having a Decentralized Infrastructure would not necessarily help you. It depends on what your goals are. Because "analytics is making it much easier to recombine new technologies in new ways", that increases the value of the fundamental new technology. Once you create it, you can quickly explore it to make new combinations. There is a tradeoff between the two.

Is there a context in which you found out that Data Analytics could impede Innovation?

Wu maintains that "he had no conclusive evidence that it prevents it, but he definitely found that the Analysis does not help build or create a novo innovation that is fundamental and can act as a future building block for future combinations. That is something in what analysts aren't good at. If you think about it, if something is that new, it probably didn't exist in the data yet. So there's not much you can do with data analysis to help you find that pattern".

Of course, recombinations can be "Radical Innovations"; they can have many profound impacts. "Many innovations are re-combinational". In this sense, I think Analytics is really moving forward on what we could do to speed up the Innovation Process.

What's Next for Wu's Innovation Research?

Wu says that we are on the cusp of a major change in technology, especially with the rise of Artificial Intelligence (AI) and Machine Learning that is dramatically changing employment and the way work is organized. Wu is looking for analytics, including the Artificial Intelligence and Deep Machine Learning subfield, to examine how we can use it more effectively to Innovate. That is an emerging problem. We can use the lessons learned, Wu concludes.

Data Driven Innovation.

Today, what organizations try to do is "to look for ways to generate Data Driven Innovation". Data Driven Innovation (DDI) is increasingly visible at different points. We can see this in the governments that have been opening data and generating innovation based on the open data trend. We also see them with tech titans like Google and Facebook, who have dedicated teams to harness data to improve their products and differentiate themselves from others in the industry. Younger companies are also organizing around "Data Science" to improve their user experiences and impact the business. In the process, companies create new tools that open up the community to power Innovation.

Another innovation in this regard is related to the fact that Data Driven Innovation refers to an innovative project born from trends or correlations in data. This project could be a new system, process, or product that has not yet been invented, but will continue to serve a measurable need or problem.

There are thousands of innovative products that use big data to function, with many more processes and systems optimized through big data. However, there is a subtle but important distinction between platforms that use data and platforms that are born from observations within a datasets. Examples:

1) Netflix's House of Card: It was one of Netflix's star shows. In 2013, 86% of their subscribers said they were less likely to unsubscribe because of this show alone. The really fascinating fact, though, is that Netflix knew the show would be a hit before it even aired. By carefully analyzing its data sets, Netflix noted that there was a correlation between fans of the original "House of Cards" British TV show and fans of Kevin Sapacey and director David Fincher. Netflix brought these three elements together in one show and voila, an instant cult classic.

2) Diapers and Beer at Walmart: Beer and baby diapers aren't two things you'd normally associate with each other. However, these two products have become infamous in data science circles, due to their unique relationship. In 1992, Teradata analyst Karen Heath found that men visiting Walmart were very likely to buy beer every time they stopped by to buy baby diapers. By placing the two items near each other at the point of sale, she was able to increase sales of both items by a significant margin.

3) Intuit: It is the team behind Quicken, a leading personal finance management tool. In the late 1990s, Quicken analysts wanted to know more about how customers were using their platform. During an analysis of geographic data, the team found, to their surprise, that more than 50% of users used the platform at their workplace. Initially, believing that customers were spending company time on their personal finances, they scheduled a few interviews with users. It turned out that the Quicken platform was so useful that many users extended the functionality of the platform to run their business accounting systems. Spotting an opportunity, the team turned the data over to the product development team, and Quickbooks was born.

Netflix and Quicken are examples of Data Driven Innovation. The reason is that these two cases are examples of new platforms, products or systems that were created due to observable data correlations. The story of diapers and beer is a perfect example of data-driven optimization of an existing system, but there is no real innovation there.

These examples illustrate an interesting dichotomy between Innovation and Optimization. While the latter is certainly essential to your business, it doesn't contain the same potential for dramatic organizational change that true innovation does. If Innovation is what you're looking for, it's wise to plan for some pretty dramatic changes within your business landscape and Culture.

There are two ways to Innovate with Database: a) Data Science (Into); b) Open Data (Outside). The organization's "Innovation Inward" refers to its being aligned with analytics, big data, and data science. The organization's "Innovation Outward" refers to being aligned with the open data and open innovation movement.

In the ideal world, organizations bring together aspects of Innovation Inward (Data Science) and Outward Innovation (Open Data and Open Innovation) in order to more efficiently generate new products and services in a Collaborative way. Netflix, for example, beyond having specialized teams to generate innovation with a database, also made the challenge in which it opened data to allow the community of data scientists to improve the results of its recommendation algorithm. In the same way, companies like Expedia generate hackathons in which developers can be part of the generation of innovative products in collaboration with companies.

Data is becoming more and more important for organizations, since the possibility of generating impact based on their exploitation is increasingly evident. For this reason, we are seeing more positions for Data Scientists around the world, as well as for leaders within organizations that have the responsibility to generate more experimentation and innovations in the database (Chief Data Officers).

Algorithms, Artificial Intelligence and Innovation.

Algorithms are understood as "a finite sequence of well-defined instructions, which is normally used to solve a class of specific problems or to perform a calculation. Algorithms are used as specifications to perform calculations, data processing, automatic reasoning, making automated decisions and other tasks".

An informal definition, could be "a set of rules that precisely defines a sequence of operations that would include all computer programs (including programs that do not perform numerical calculations) and, for example, any prescribed bureaucratic procedure or cookbook recipe" .

In general, a program is only an algorithm if it eventually stops, although infinite loops can sometimes be desirable.

As an effective method, an algorithm can be expressed within a finite amount of space and time, and in a well-defined formal language to compute a function. Starting from an initial state and an initial input, the instructions describe a computation that, when executed, progresses through a finite number of well-defined successive states, eventually producing an output and ending in a final state. The transition from one state to the next is not necessarily deterministic; some algorithms, known as random algorithms, incorporate random inputs.

Most algorithms are intended to be implemented as computer programs. However, algorithms are also implemented by other means, such as "a biological neural network" (for example, the human brain implementing arithmetic or an insect searching for food), in an electrical circuit, or in a mechanical device.

Without going into depth about Artificial Intelligence , I will say it refers to "the intelligence expressed by machines, their processors and their software that would be analogous to the body, the brain and the mind, respectively, unlike the Natural Intelligence demonstrated by humans and certain animals with complex brains".

People talk about Artificial Intelligence in machines, colloquially, when it mimics the cognitive functions that humans associate with other human minds, such as "perceiving, reasoning, learning and solving problems".

Andreas Kaplan and Michael Haenlen define Artificial Intelligence as "the ability of a system to correctly interpret external data, to learn from that data, and to use that knowledge to accomplish specific tasks and goals through flexible adaptation". Artificial Intelligence is a new way of solving problems that includes expert systems, robot management and control, and processors that try to integrate knowledge into such systems, in other words, an intelligent system capable of writing its own program. An "Expert System" is defined as a programming structure capable of storing and using knowledge about a certain area that translates into its learning capacity. Similarly, Artificial Intelligence can be considered as "the ability of machines to use algorithms, learn from data and use what they have learned in making decisions just as a human being would, in addition to one of the approaches principles of artificial intelligence that is Machine Learning, in such a way that computers or machines have the ability to learn without being programmed to do so".

Thus, today, the applications of Artificial Intelligence range from computational linguistics, data mining or Data Mining, Industry, Medicine, Virtual Worlds, Natural Language Processing, Robotics, Control Systems, Decision Support Systems, Video games, Computer Prototypes, Dynamic System Analysis, Crowd Simulation, Operating Systems, Automotive, among others.

Thus, the idea that Artificial Intelligence and Machine Learning can replace humans, take over roles in the workplace and reshape existing organizational processes, has been growing steadily (Brynjolfsson, McAfee, 2017; von Krogh, 2018). The central premise is that, given certain limitations, artificial intelligence can provide higher quality, greater efficiency and better results than human experts.

Considering the potential of Artificial Intelligence to take over traditional human tasks in organizations, we may wonder if a role can be used for Artificial Intelligence in pursuit of one of the most important processes affecting long-term survival and advantage. of organizations: Innovation (Lengnick-Hall, 1992; Porter & Sterm, 2001). The idea prevails that Artificial Intelligence and Machine Learning could and should be used by companies for Innovation purposes, which may seem almost unreasonable. After all, Innovation has traditionally been seen as the domain of humans, given their "unique" ability to be Innovators (Amabile, 2019).

Although Artificial Intelligence may have disadvantages compared to humans, there are several non-trivial reasons why companies may want to use Artificial Intelligence in their Innovation processes. Among the exogenous factors to the Innovation Process, is the fact that "innovation managers are increasingly faced with highly volatile and changing environments, increasingly competitive global markets, rival technologies and drastically changing political landscapes" (Jones, 2016; O′Cass & Wetzels, 2018; Spieth, 2014). At the same time, the availability of information has increased and continues to increase significantly. These trends provide strong evidence that the foundation for competitiveness is based on the information and problem-solving capabilities of organizations (Hajli & Featherman, 2018). Perhaps most importantly, in many areas, the negative effects of Innovation risk are compounded by rising costs. That is, "the cost of each innovation has been increasing quite dramatically". For example, in the pharmaceutical industry (Munos, 2009; Pammolli, 2011). This means that the way Innovation is organized needs to be challenged by introducing Artificial Intelligence and Machine Learning due to their cost advantages in information processing.

Therefore, finding ways to apply Artificial Intelligence and Machine Learning to the Innovation Processes of companies should be of great interest to Innovation managers. On the one hand, this has the potential to create better ways for companies to respond to their increasingly competitive environment and manage the growing amount of information that surrounds them. On the other hand, supporting the Innovation Process with Artificial Intelligence could generate real value for companies by reducing both the risk and the cost of the Innovation Processes.

Potential Application Areas of Artificial Intelligence in the Innovation Process.

By combining the barriers that both humans and Artificial Intelligence systems need to overcome in the Innovation Process with the key idea generation and development activities that need to be carried out, we can derive a framework of potentially creative Artificial intelligence application areas. within the Innovation Process. Artificial Intelligence can help and potentially replace human decision-making in Innovation Management, specifically in the following potential areas (in which it could support human decision-making): a) Develop Ideas Overcoming Information Processing Limitations , b) Generating Ideas Overcoming Information Processing Limitations, c) Developing Ideas Overcoming Local Search Routines, d) Generating Ideas Overcoming Local Search Routines.

Overcome Information Processing Limitations with Artificial Intelligence to Develop Ideas: There are many exciting applications of artificial intelligence systems in material discovery. For example, Artificial Intelligence can be used to optimize battery components and solar cells, or to speed up the discovery process for new catalysts. To discover these new materials, Machine Learning-based methods are used to predict the most promising materials to test, substantially accelerating the Innovation Process. There are Artificial Intelligence systems for pharmaceutical research and development, including uses that accelerate the Protein Engineering Process, which is essential to discover suitable proteins for technological, scientific and medical applications. Artificial Intelligence applications can be used to identify treatments for diseases; for example, deep-domain adaptive neural networks have been trained on single-cell RNA genomic datasets to develop treatments that stop Malaria transmission. There are other applications as well, for example Celonis uses process mining to identify organizational processes that are suitable for robotic process automation. Celonis uses Artificial Intelligence applications that allow organizations to implement important administrative innovations.

Overcome Information Processing Limitations with Artificial Intelligence to Generate Ideas: There are Artificial Intelligence applications that can process much more information to generate new ideas and opportunities that would likely be missed by humans operating on their own. For example, Outlier.ai . The company uses a set of Machine Learning methods to Process raw Metrics Data and turn it into human-readable knowledge. After analyzing a company's data, Outlier generates a set of personalized "stories" that summarize actionable and interesting information for specific managers. By doing so, Outlier can highlight innovative opportunities for managers. Thus, the analysis based on Artificial Intelligence provided by Outlier was essential to develop an innovation in the focal company. Outlier's ability to find significant anomalies and patterns in business data is one way Artificial Intelligence can help companies generate or recognize innovative ideas and opportunities. These Artificial Intelligence methods may not be able to independently develop complete solutions, but they can guide human managers towards the most promising avenues for Innovation.

Another example is provided by Tshitoyan and his colleagues. They created an Artificial Intelligence system that can capture latent knowledge from the materials science literature. Their system uses the word2vec algorithm, a popular neural network in natural language processing applications, to derive embeddings of concepts in the literature. The algorithm is capable of capturing complex materials science concepts, including the underlying structure of the periodic table, without any explicit input of chemical knowledge by researchers. The Artificial Intelligence system can also recommend materials for functional applications. By censoring the data, the authors can show that the system is, in fact, capable of recommending materials several years before their discovery. Therefore, this method points to potential opportunities for future innovations, albeit within an already existing domain of knowledge. This study is indicative of possible applications of Artificial Intelligence, that is, Artificial Intelligence systems that can help to generate or recognize ideas and innovation opportunities where a large amount of information must be processed in an existing knowledge domain.

Overcome Local Search Routines with Artificial Intelligence to Develop Ideas: These activities involve identifying and developing ideas, opportunities and solution approaches where the process goes beyond the use of local search routines, i.e. remote search is used. Autodesk , for example, used various algorithms to create a new crew partition for Airbus. The generative design methods used to design the new partition create the kinds of products that the designers couldn't conjure up on their own. The algorithms used by Autodesk were based on the growth patterns of slime mold and mammalian bone. They allowed the construction of a new, more efficient, but equally stable crew partition. Therefore, by incorporating Artificial Intelligence methods into the Development Process, Autodesk and Airbus were able to generate a more innovative solution than would have been possible otherwise.

Even more interesting are the applications based on Generative Adversarial Networks (GAN). The Creative Confrontation Network (CAN) for art creation developed by Elgammal and his colleagues is an example of such an Artificial Intelligence solution. The CAN is a type of GAN that is capable of generating novel art. The network is trained on 81,449 paintings by 1,119 artists ranging from the 15th to the 20th century. The system trains two competing networks, a "discriminator" and a "generator" to learn art style classification (discriminator) and style ambiguity (generator). As a result, CAN generates new art that deviates from learned styles. We would argue that this departure from previously learned styles is precisely where the CAN system is able to outperform local search routines and show its potential for remote searching. Since the model initially learns about existing art styles, it has knowledge about the current domain. However, it is set up to specifically explore beyond current styles and is therefore capable of generating novel ideas. Another related research project of Sbai and his colleagues is called DesIGN: Design Inspiration for Generative Networks. This system can generate novel styles, shapes and forms for fashion garments. DesIGN deviates from existing fashion styles as represented in the training dataset, while generating realistic clothing items. Therefore, it surpasses local search routines when developing new ideas for fashionable garments.

Overcome Local Search Routines with Artificial Intelligence to Generate Ideas: Finally, the Artificial Intelligence systems here must be able to generate or recognize ideas and opportunities for innovation in unrelated domains of knowledge. A method in Artificial Intelligence that can facilitate the generation or recognition of innovative ideas and opportunities is Reinforcement Learning. There have been recent advances in Reinforcement Learning, such as Meta-Reinforcement Learning, which could possibly be useful in generating novel ideas. Reinforcement Learning, in general, involves training an agent in a (virtual) environment. The agent uses a reward signal to know which actions maximize rewards and which ones decrease them. Reinforcement Learning requires humans to craft rewards by hand, which is a non-trivial and sometimes suboptimal approach to reward engineering. As Simon Osindero, one of Google DeepMind's leading AI researchers, explains:

"To the extent that you're hand-designing a reward function, in a sense you're also hand-designing a solution. If it were easy for us to design a solution, then maybe you wouldn't need to learn in the first place".

Unsupervised Reinforcement Learning attempts to address this deficiency by allowing the agent to learn its reward function through a stream of observations and actions. Therefore, this method is a first step to allow algorithms to learn to recognize and achieve goals without any supervision, which will open interesting avenues for Creativity and Innovation. Meta-Reinforcement Learning addresses a closely related question of how Learning can be used to improve the Learning Process itself. Recent work in this area has attempted to design algorithms that can be rapidly adapted to arbitrary new problems. Advances in these areas should allow algorithms to become more flexible in terms of solving new problems, which can be helpful in generating, discovering, and recognizing new ideas and creative opportunities.

Conclusions.

Innovation Management can be supported by Artificial Intelligence systems. Conventional human-centric approaches to Innovation Management have limitations that stem primarily from their imperfect ability to fully address information needs and cope with complexity.

Information Processing restrictions derive in levels of information processing capacity of Artificial Intelligence necessary to develop digitized organizations. Finally, it is possible to outline the challenges in the implementation of Artificial Intelligence systems that Innovation Management faces in relation to the technology itself, the humans in charge of implementing it and the technology-human nexus. Thus, Artificial Intelligence has a constructive role to play when the tried and true benefits of Innovation Management resources are overwhelmed, made impossible due to digitization or when Artificial Intelligence irrefutably emerges as the preferred option.

It seems clear that the potential of Artificial Intelligence lies in "creating a more systematic approach by integrating Artificial Intelligence in organizations that pursue Innovation". Innovation Management sheds light on the use of Artificial Intelligence and Machine Learning algorithms in the future organization of Innovation. The areas in which Artificial Intelligence Systems can already be fruitfully applied in Organizational Innovation, that is, instances in which the development of new innovations is mainly hampered by restrictions in information processing. Artificial Intelligence systems that are based on anomaly detection, for example, can be useful when companies face limitations in information processing while searching for new opportunities.

Finally, there are recent advances in Artificial Intelligence algorithms that are indicative of the potential of Artificial Intelligence to solve the most difficult challenges in Innovation Management. These include beating local search and generating entirely new ideas. Thus, Artificial Intelligence will open up new possibilities and expand the areas in which Artificial Intelligence can be usefully applied in Innovation Management.

Therefore, answering the initial question about whether "algorithms can Innovate", the answer would be: "Apparently they are already doing it".

(Note: For those students of the course, go back to the course and follow the instructions to assimilate the knowledge delivered).

Source: "Artificial Intelligence and Innovation Management: A Review, Framework, and Research Agenda", Naomi Haefner, Joakim Wincent, Vimit Parida, Oliver Gassmann, Technological Forecasting and Social Change; "Ejemplos de Inteligencia Artificial en la Actualidad", Transformación Digital, Beetrack; "How Data Analytics Can Drive Innovation", Knowledge@Wharton, Wharton University of Pennsylvania; "Innovando con Datos: Datos Abiertos y Ciencia de Datos", Diego May, IADB; "Real Life Example of True Data Driven Innovation", Owen Hunnam, Idea Drop;