Close
< Back to Perspectives Home

AI and the ESA: Continuing a Conversation

This is Part II of a two-part series about AI and the ESA

By Thomas Collins and Susan Leddick

“AI will be economically significant precisely because it will make something important much cheaper. What will AI technologies make so cheap? Prediction.”

“Organizations can exploit prediction machines by adopting AI tools to assist with executing current strategy. When those tools become powerful, they may motivate changing the strategy itself.”
Prediction Machines: The Simple Economics of Artificial Intelligence (Ajay Agrawal, 2018)

Introduction

In the October issue of AESA Perspectives, we invited you to become part of a conversation to continue at AESA conferences, in your leadership team meetings, and with ESAs in your region and state on the strategic implications of artificial intelligence (AI) and machine learning (ML) for ESAs. This month, we would like to expand and deepen aspects of the conversation.

Among our key points in October were, generally, AI deals with technologies, systems or processes that competently mimic how human beings react to new information, speak, hear, understand language, and make predictions. AI and its variants, including especially ML, require us as ESA leaders to re-evaluate the competitive forces we face, the competitive strategies we employ, and the ways we develop and deliver services. Despite the obvious connection between how AI will affect education systems as a whole, we have focused on implications for ESAs in these two articles.  We know that ESAs and the education system are inseparable in the long run, but a focus on ESAs have dominated our thoughts.

There is plenty of hype regarding AI, and Rand researcher Robert Murphy refers to an “evidence gap” (Murphy, 2019), but there is also plenty of substance. Our intent in this article is to explore specific implications of AI for the core business and operations of ESAs. We will suggest specific and concrete steps your ESA can be taking right now to better understand and prepare for AI-related risks and opportunities. We will also offer guidance on experimenting with AI projects.

A Useful Thinking Framework

Dr. Duncan Simester, Professor of Management at MIT Sloan, has introduced many ESAs to the Business Strategy Framework (BSF). We find this to be a useful organizer to help ESAs to explore the strategic implications of AI for their organizations, and we believe that a systematic strategic approach is essential to analyzing the impact of AI on ESAs. Some ESAs are familiar with the framework Dr. Simester teaches and referencing it provides us a common language including terms such as value, horizontal competitor, vertical competitor, and value chain. Even those ESA leaders who have not been directly exposed to the framework can still use its basic ideas productively. More on the BSF can be found in the Appendix.

Strategic Implications for the Core Business of ESAs

The First Wave

In Karen Cerulo’s book Never Saw It Coming (Cerulo, 2006) she describes what she believes to be a uniquely American tendency to focus on and exaggerate the best-case and most optimistic outcomes, and to undervalue worst-case scenarios. Looking through our rose-colored glasses, we can be shocked when less desirable outcomes and consequences occur because we never saw them coming.

We CAN and DO see AI coming, however. The global education market for AI, valued at approximately $265M in 2016, is anticipated to grow at a compound growth rate of more than 45% over the forecast period 2018-2025 (Artificial Intelligence Markets in the US Education Sector 2018-2022).

As we established in first article, prediction problems to which machine learning (ML) and predictive analytics are applied are the first wave of AI products to be targeted to the education market, including ESAs. Predictive analytics use statistical models and forecasting techniques to understand the future and answer the question, “What could happen?”  (Descriptive, Predictive, and Prescriptive Analytics Explained, 2019). We recommend that ESAs focus initial efforts on prediction (what could happen), not prescription (what should we do).  Examples that are already or soon will be in the marketplace include personalized learning, dropout/failure prevention, depression and suicide prevention, pattern recognition in speech and many others. Some providers of products like these have already begun to partner with ESAs in several states, but   ESAs should expect that AI/ML developers and providers will sell directly to schools and districts, inserting themselves between your ESA and its clients. For example, one learning management system (LMS) touts, Our in-house service teams include: Implementation, Training, Learning Strategy and Consulting, Learning and Creative Services, Data Services and Support. Our education experts can help you dig deeper and reach higher, to help you deliver simply the best learning experience for your students, teachers, and employees. Your success is our priority. Sounds like something an ESA might say, right? So, what is an ESA to do?

Business Model Implications

Build Your Own?  Maybe…

At this point, you may be asking yourself, “Can our ESA, or a group of ESAs, develop its own AI tools? Can we harness AI technology for ourselves and our customers?” Until recently, the answer was probably not, for several reasons. First, developing AI tools in-house requires expensive personnel with specialized knowledge and skill sets, typically unavailable or unaffordable for most ESAs. Second, in addition to programming expertise, developing AI tools requires dedicated content-area specialists to work with programmers. That type of excess capacity is typically unavailable or unaffordable for most ESAs. Third, successful AI projects require huge amounts of data of three types: training data used to develop the specific tool, input data used to make a prediction, and feedback data used to improve the tool’s effectiveness over time. Most individual ESAs are too small to have access to that amount of data, though possibilities may exist across multiple ESAs and at the state department of education level. Training data often requires labeling, and again ESAs are not likely to have the wherewithal to manage that task, either. Finally, many AI-focused companies are well financed and at or approaching scale right now. Successful smaller companies and their talent pools are often bought by larger companies. As of August 2019, there have already been over 140 such acquisitions, putting the year on track to beat the 2018 record (The Race For AI: Here Are The Tech Giants Rushing To Snap Up Artificial Intelligence Startups, 2019). Leading the list are firms like Apple, Google, Microsoft, Facebook and Amazon. Clearly the barriers to entry as a producer for this market are very high for ESAs.

What’s Ahead for Competition?

A more fundamental challenge to “building it ourselves” is that nearly every significant AI-related horizontal competitor to ESAs operates on a fundamentally different business model that will make them formidable competitors. Consider the following quotation:

For most of the last hundred years, every company competed on the same economic playing field—make things and then sell those things and try to grow to take advantage of scale. The same with services.  Hire some people, bill some hours, hire more people. However, in the last thirty years, new business models, those that leverage scalable technology, data and networks have turned those old-world business models on their head.  Once premier businesses and brands are now struggling legacy organizations with eroding revenues, profits and value…. The problem is that leaders have long focused on making and selling things and offering the services of people, the least valuable business models. (Beck, 2019)

Nationwide, there are significant differences among ESAs and many have evolved or are evolving their own business models. That said, it is still safe to say that many if not most ESA business models are in Beck’s second category—predominantly people-intensive and difficult and costly to scale even assuming qualified people and adequate resources are available.

Is all lost? Are ESAs doomed by high-resource, AI-enhanced competitors? We do not believe so, but we are convinced that ESAs must take a hard look at how best to compete in this new market —and the sooner the better. So, how to begin? Experiment. Though most ESAs probably cannot build from-the-ground-up AI applications themselves, AI Super-Powers: China, Silicon Valley and the New Work Order (Lee, 2018) describes the emergence of cloud-based “commoditized AI” platforms available by subscription. Microsoft Azure, AWS Forecast, AWS Deep Learning, and IBM Watson are examples of this emerging approach. These companies, and others, make AI development tools more accessible, though many of the data and capacity challenges likely will remain for less technologically sophisticated or data-challenged ESA.

What Can ESAs Do Now, and What Are Some Already Doing?

An essential challenge is choosing potential problems to solve that are within the ESA context. As noted, large companies with significant resources are targeting the K-16 education marketplace. There is probably not much opportunity for ESAs in that segment of the market. However, ESAs occupy a unique space with potential to develop niche products and services. As your ESA considers problem selection, we recommend starting with a focus on prediction problems, seeking the lowest-hanging fruit and single short-term outcomes. This will require close examination of the data you have or to which you have access: staff, student, academic, discipline, financial, human resource, and more. While this may seem obvious, we further recommend that you start with the intent to create something of value for your ESA and/or your clients. Supply side AI applications can help you do your ESA business better or more cheaply. Demand side AI application can help your clients do some aspect of their business better or more cheaply than they now can. If it is a demand side application, consider which market segments your application is intended to serve (students, teachers, parents, administrators, others). Additional guidance on this process can be found here and here.

Some ESA AI-driven applications are beginning to emerge. As an example, Iowa AEAs have worked as a statewide network of all nine agencies to develop and launch Scout, which uses a Google-like interface and an Amazon-like recommendation engine (“If you like THIS resource, you might like THESE resources, too.”) for its online portal to digital resources provided by the AEAs for students and teachers. More about Scout can be found here. As another example, still in the exploratory phase, an ESC in Texas is actively considering how to use AWS Forecast and similar tools with what it considers to be a unique strategic resource that it has, namely, years of data from its schools. This ESA is not yet certain what outcomes it may target, but it is actively exploring the question and considering possible experiments to determine what it can learn.

As above, we recommend that your ESA avoid prescription problems (“What should we do?”) for several reasons: this type of problem requires an enormous amount of training data and the output its algorithms produce are not fully explainable. The inability to explain how the algorithm does what it does opens up a host of potential legal and ethical concerns that we believe are best avoided until “algorithm explainability” improves. In addition, this aspect of the current technology in ML and AI invites skepticism and distrust of the recommendations the software makes. ESAs should be aware that not all staff nor all practitioners will immediately embrace AI-engineered solutions.

After problem selection, the next consideration is choosing a platform (Watson, Azure, AWS, and others) and appropriate algorithm(s).

The answer to the question, “What machine learning algorithm should I use?” is always “It depends.” Variables include the size, quality, and nature of the data. It depends on what you want to do with the answer. It depends on how the math of the algorithm was translated into instructions for the computer you are using. And it depends on how much time you have. Even the most experienced data scientists can’t tell which algorithm will perform best before trying them.” (How to Choose Algorithms for Azure Machine Learning Studio, 2019).

The Microsoft Azure site offers guidance on algorithm selection.

To summarize, there are significant challenges ahead for an ESA that wants to experiment with AI, but we also know that some ESAs are taking up and will take up the challenge. As with any type of experiment, expect to fail, often! That is why we experiment, to avoid the big miss, to avoid the significant potential consequences of failing at scale. Wherever you maybe on your AI journey, we invite you to share your learning with us and others.

What Are Implications for the ESA Value Chain?

We believe that AI-enhanced products and services will disrupt traditional ESA value chains by introducing not just new horizontal competitors who will offer AI-enhanced solutions directly to schools and districts, but also new vertical competitors who represent options for ESA partnerships and new service development. Competing with formidable horizontal AI competitors will be a challenge, but we believe that the opportunities for ESAs are two-fold and offer a counterbalance to the threat. First is the opportunity to strike productive partnerships with vertical competitors such as AI and ML software providers. It will be essential to identify what the ESA has access to or can do to add value that other actors within the value chain cannot. Offering technical assistance to makers of a product such as PowerSchool is a well-established ESA practice that is easily transferrable to AI product producers. The Texas ESC mentioned above believes it has a unique strategic resource in its access to years of student data from its districts. An ESA’s unique contribution may vary in different situations and service areas: one time as a problem identifier, another as pilot or test sponsor, and another as after-sale technical assistance provider, for example. Some of these roles are more unconventional for ESAs than others.

A second counterbalancing opportunity is to develop and offer new products and services.  Thinking again of the steps in the value chain, we could ask how ESAs might play a role in problem identification on the front end of a product development process. How might ESAs serve as advanced scouts for their clients, researching potentially valuable AI-enhanced products and services? How might ESAs develop internal staff capacity to support applications that are in use within school districts in their regions? Post-sale, we believe that schools and districts will have a strong need for technical assistance in using the information provided by the various products. We believe technical assistance is different from technical support, which will more likely come from the companies, themselves. The relevant players may be different for each potential customer group such as decision-makers, implementers, or users. Finding the optimal place for ESAs should take both the value chain and the customer segment into account.

Examining vertical competition within the value chain can expose additional risks for an ESA, in addition to the danger of being cut out of the chain altogether. Identify risk to the ESA by asking, “What kinds of problems that are traditionally addressed by ESAs may be more effectively addressed by AI-enhanced products?” For instance, how will the work of specialized ESA consultants who perform data analysis, problem identification, intervention design, and professional development be disrupted when smart products can perform all or most of those consulting tasks? How will the role of an ESA speech pathologist be impacted by an AI-enhanced subscription-based product that can quickly and accurately diagnose student speech issues and, soon, offer prescriptive remediation?

Summarizing some of the value chain implications, partnerships defined in contracts and new service opportunities are two typical ways for ESAs to reduce the power of vertical competitors and to avoid being completely squeezed out.  Partnerships rely on an exchange of value, though, and each partner must clearly spell out its contribution. No contribution by the ESA equals no meaningful partnership. The partnership may mean sharing work, resources, and revenue, all of which require clear contractual definition.

Jobs and Services Will Evolve

AI will be used to streamline and enhance operational efficiencies at the ESA and in the schools and districts it serves. In our first article, we suggested that to remain cost-competitive, ESAs need to examine HR, financial, and other processes that may be personnel-intensive or expensive to operate and seek to complement those processes with new tech tools. We concluded that AI is unlikely to replace “whole jobs” and “whole people” but instead is more likely to replace certain functions within jobs, creating hybrid jobs where people perform functions at which humans are best and where smart machines do what smart machines do best. When thinking about the potential impact of AI, it is helpful to think not of individual employees but of the bundles of tasks or functions that, when performed over the course of a day or week, constitute that employee’s job. HR and Finance are among the first places many organizations seem to be looking to identify internal AI-enhanced solutions. The tasks most likely to be automated first are the ones for which automation delivers the most ROI (Ajay Agrawal, 2018). Candidates for AI-enhanced solutions are jobs with tasks characterized by a high degree of structure and repeatability, little or no direct contact with customers, heavy use of quantifiable data or codifiable knowledge, entry-level skills, and no direct generation of revenue or profit.

New technologies also offer opportunities for new ways of looking at the ESA business and potentially new services. We’ve mentioned Lexio that helps turn spreadsheets and other data sources into narratives and stories more easily understood by non-expert data consumers. How might ESAs use such a tool to convert, or complement, their annual reports to narratives without losing rigor?  And how might the use of such tools increase the effectiveness of communication services that ESAs already offer to their members?  AESA business partner Forecast 5 says it offers an alternative to traditional spreadsheet-based forecasting methods. Notice in Forecast 5’s offerings the emphasis on prediction problems—the sweet spot for AI and ML applications.  Many of the examples Forecast 5 cites are precisely the kinds of tasks ESAs and school districts routinely perform.  ESAs may use such products both in their own operations and to improve the services they offer.

We also believe that with AI and ML tools at their disposal, ESAs will be able to offer valuable new services by assisting districts with district financials, transportation routing, facilities maintenance, and other similar processes. Much of the relevant financial, transportation, maintenance, and other operational data is available now and can be accessed with the right permissions and appropriate data privacy agreements.

We believe there is only one way for ESA-developed demand-side products or services to be successful, whether AI-based or other. Identify a niche customer market, then fill it. If the niche disappears, is commoditized, or the ESA can no longer compete, get out of that market and seek new niches. For an ESA product or service to be successful, it must provide the ESA a sustainable differentiated competitive advantage. By “sustainable” we mean that it provides ongoing value to the ESA and its customers in the face of persistent investment by competitors trying to replicate it. As ESAs consider the development of AI-related products and services, the same guidance applies: if you cannot identify a sustainable differentiated competitive advantage, rethink your strategy.

Additional Considerations

In whichever ways an ESA chooses to engage with AI, data governance is an essential consideration. Current ESA data policies and procedures likely are aligned to state and federal regulations. A potential new data policy consideration is mitigation of potential AI-related biases and discrimination (Goasduff, 2019).

AI needs data, more than that to which a typical ESA has access, notwithstanding the examples of operational data we mentioned above. How might sharing data across state ESA networks and across districts help reach critical thresholds to assure the quantity and quality of data requisite for AI to be effective?  Of course, data sharing among ESAs, districts, and others has cultural, structural, and legal hurdles to overcome, as some in the ESA community may recall discovering during the national AESA-sponsored initiative in 2012 and after to gather some standardized ESA performance measures. The working group found that ESA data measurement systems were ill-defined, highly variable agency-to-agency, and that there was not insubstantial reluctance to share any data that could reflect poorly on an ESA. Further, data ownership and data privacy and protection are essential aspects to be considered. This is especially important at the local and regional level which may not yet have adequate data privacy policies in force, but may be more adequately addressed at a state department of education levels where data protection measures are typically in place. These challenges must be addressed if ESAs, individually and collectively, are to be relevant players in the AI future.

How ready is the ESA to operate in an environment in which there is much higher reliance on AI-enhanced solutions to clients’ problems?  How may the organization develop ESA staff competence to support AI applications in client districts as well inside the ESA? What processes will ESAs establish to maintain contextual awareness to inform strategic decisions about which services are best performed by the ESA and which not? The answers to these important questions imply keeping up with a rapidly changing AI/ML landscape and keeping track of which products are being used within the ESA customer base. Beyond knowing what’s going on in the region, an ESA’s AI strategy will hinge on understanding which decision criteria are most important to a given client or group of clients and dispassionately comparing the agency’s ability to differentiate its services from those of competitors as well as evaluating the agency’s resources in the service area in question.  Especially intriguing are niche markets, narrow service opportunities where the ESA can add value quickly to client AI and ML users, all the while being ready to exit and find another niche opportunity as specific problems are solved or competition changes.

Partnerships are tempting. Companies dangle shared-revenue carrots that can be very appealing—especially in times of shrinking public resources for ESAs.  How can an ESA be sure that an AI “solution” really addresses a valid problem before considering any partnerships with vendors?  How big a problem is it?  What’s the likelihood that the company’s “solution” will really solve it? How shall we evaluate partnership agreements and renegotiate them over time as value gets re-defined and contributions shift in the value chain?  Many an ESA has been cut out of a value chain because it agreed to be a re-seller without considering how to add, maintain, and protect its own value.  As demonstrated in AESA’s Jumpstart meeting in February of 2014 between ESA leaders and owners of start-ups in education, ESAs walk a fine line to be perceived by their school districts as a service provider, not a vendor. In addition, partnerships demand purposeful management and take up valuable time of key staff to do so. Partnerships are always a balancing act between how much to cooperate and how much to share revenue.

Clearly, there are many more questions about the strategic implications of AI and ML in ESAs than answers. That is why we believe conversation is so important right now before we lose more valuable time.

Getting Started

Thinking before acting is generally good practice but not always generally practiced well.  We have suggested the concepts of the Business Strategy Framework (BSF) as a helpful tool for evaluating the strategic implications of AI and ML in your ESA and its service offerings. Using Business Strategy Framework concepts such as horizontal competitors, vertical competitors and value chain, and value for which money is exchanged in a structured thinking process before committing your organization’s resources can be very useful. Use cases can help you and your leadership team clarify your thinking about AI and ML. The keys are specificity, manageability, examining multiple variables, and understanding risk and opportunity for your ESA. Look at the situation from several angles: different customer groups, different customer search behavior and decision criteria, different competitors, different places to contribute to the value chain.

Making AI a regular element of strategic conversations in your agency is probably the most critical single step to take toward preparing for AI and ML.  As we have discussed, we believe AI and its variants are here to stay and will have important strategic implications for ESAs. We recommend that ESAs give AI a heightened level of strategic importance and begin to implement practices such as the following:

  • Establish a learning protocol at the cabinet-level to inform your leaders about what is happening in AI for education and ESAs and to identify how districts in your state or region are beginning to incorporate AI and ML products into their settings.
  • Extend this protocol to your state’s ESA network to broaden perspective and to identify opportunities for mutual learning, cooperation, synergy, and partnerships.
  • Connect with early ESA implementers such as Iowa’s AEA network as it launches “Scout,” a statewide resource-sharing, smart media and technology portal.  Scout incorporates user curation and prediction. Hamilton County ESC in Cincinnati is experimenting with robots with some of its autism students and is also exploring recommendation algorithms for its website. We are sure there are many other examples—and we are eager to hear about them.
  • Evaluate service opportunities that AI and ML applications offer to the ESA. Debate the ways in which your ESA can add value to the value chain for AI and ML products.
  • Begin to develop a cadre of professional staff who are interested in expanding their knowledge and skills for supporting AI applications in client districts.

Conclusion and Call to Action

In our two articles, we have tried to make the case that AI is here and accelerating in importance for ESAs. We know for certain that serious dollars are being invested in the AI-for-education market. We know for certain that AI is on your customers’ radar. We know for certain that AI will impact ESA business models in foreseeable and, importantly, unforeseeable ways. We have offered Dr. Duncan Simester’s Business Strategy Framework as a potential way to structure your analysis process and to develop a common language for conversation.

We have asked you to consider many questions, probably too many! We acknowledge that we do not, nor does anyone else, have all of the answers. That is why we have issued the invitation to join the professional conversation about AI and ESAs.  We look forward to being part of the dialogue and would like to hear about what you are thinking as well as your efforts and initiatives.

Special thanks to Daniel Hanrahan, CEO of Cooperative Educational Service Agency 2 in Whitewater, WI, and Nick Brown, Deputy Executive Director of Region 12 ESC in Waco, TX, who served as critical readers and for their review, comments, and insights.

Appendix

Key Elements of the Business Strategy Framework (BSF):

  • Horizontal competitors are defined as organizations that offer similar services to yours to the same clients or customers you serve or could serve.
  • The value chain is defined as the full sequence of activities needed to create and deliver a product or service.
  • Vertical competitors are defined as actors in the value chain who have contributed to the development and delivery of a product or service and who want to take a slice of the revenue pie or extract some other benefit for themselves, thus reducing the amount of revenue or value available for other contributors.
  • Value can be added to a product or service by saving the customer money or by improving the customer experience, or both.

A simple example takes the definitions a step farther: say I want to purchase a pair of cotton socks. The value chain includes the distributor who sells the cotton seed, the farmer who grows and harvests the cotton, the mill that transforms raw cotton into fiber, the hosiery manufacturer that transforms cotton fiber into socks, the trucking company that hauls the socks from the manufacturer to my local store, and the store that sells me the socks. Each needs to extract money from the value chain to stay in business (seed company, farmer, mill, hosiery company, etc.). To the extent that each competes with the other for a share of the value, these are vertical competitors to each other. There are also horizontal competitors to my local retailer that also would like to sell cotton socks to me—other local retail stores and online sellers, to name two. If I decide on nylon socks, though, the value chain is entirely different. The seed seller, the farmer, and the mill will be replaced in the value chain by a chemical company and its suppliers who produce the synthetic material.

References

Ajay Agrawal, J. G. (2018). Prediction Machines: The Simple Economics of Artificial Intelligence. Boston, MA: Harvard Business Review Press.

(n.d.). Artificial Intelligence Markets in the US Education Sector 2018-2022. Dublin: Research and Markets.

Beck, B. L. (2019, June 16). Only A Few Strategy Decisions Separate Good Companies From Great Ones. Retrieved from AI Matters: https://www.aimatters.com/blogs-and-articles/2019/6/21/only-a-few-strategy-decisions-separate-good-companies-from-great-ones

Cerulo, K. A. (2006). Never Saw It Coming: Cultural Challenges to Envisioning the Worst. Chicago and London: The University of Chicago Press.

Columbus, L. (2018, January 12). 10 Charts That Will Change Your Perspective On Artificial Intelligence’s Growth. Retrieved from Forbes: https://www.forbes.com/sites/louiscolumbus/2018/01/12/10-charts-that-will-change-your-perspective-on-artificial-intelligences-growth/#1a020d9f4758

Davenport, T. H. (2018). The AI Advantage: How to Put the Artificial Intelligence Revolution to Work. Boston: MIT Press.

David Autor, D. A. (2019). The Work of the Future: Shaping Technology and Institutions. Boston: MIT .

David Schatsky, C. M. (2015, January 27). Cognitive Technologies: The Real Opportunties for Business. Deloitte Review Issue 16.

Descriptive, Predictive, and Prescriptive Analytics Explained. (2019). Retrieved from Halo: https://halobi.com/blog/descriptive-predictive-and-prescriptive-analytics-explained/

Global AI in Education Market 2019 by Technology, Application, Solutions, Services, Emerging Trends, End Use and Investment Opportunities to 2025. (2019, April 27). Retrieved from Reuters Plus: https://www.reuters.com/brandfeatures/venture-capital/article?id=104547

Goasduff, L. (2019, September 12). Top Trends on the Gartner Hype Cycle for Artificial Intelligence, 2019. Retrieved from Gardner: Smarter With Gardner.

Hardesty, L. (2016, September 22). MIT News. Retrieved from http://news.mit.edu/2016/automated-screening-childhood-communication-disorders-0922

Hepplemann, M. E. (2019). Why Every Organization Needs an Augmented Reality Strategy. HBR’s 10 Must Reads: On Analytics and the New Machine Age, 53-76.

How to Choose Algorithms for Azure Machine Learning Studio. (2019, March 3). Retrieved from Microsoft Azure: https://docs.microsoft.com/en-us/azure/machine-learning/studio/algorithm-choice

Learning @ Scale. (2019). Retrieved from Learning @ Scale: https://learningatscale.acm.org/las2019/

Lee, K.-F. (2018). AI Super-Powers: China, Silicon Valley and the New World Order. New York: Houghton Mifflin Harcourt Publishing Company.

Levitt, T. (1960). Marketing Myopia. Best of HBR, 138-49.

McKinsey Global Institute. (2017). Harnessing Automation for a Future That Works. McKinsey & Company.

MIT Faculty. (n.d.). Retrieved from MIT Management Sloan School: https://mitsloan.mit.edu/faculty/directory/duncan-simester

Murphy, R. F. (2019). Artificial Intelligence Applications to Support K–12 Teachers and Teaching. Global: Rand Corporation.

Porter, M. E. (2011). What is Strategy? In HBR’s 10 Must Reads: On Strategy (pp. 1-38). Boston, MA: Harvard University Press.

Porter, M. E. (2011). What is Strategy? In HBR’s 10 Must Reads: On Strategy (pp. 1-38). Boston: Harvard Business Review Press.

Restrepo, D. A. (2018). Artificial Intelligence and Work Working Paper 24196. Cambridge, MA: National Bureau of Economic Research.

Rouse, M. (n.d.). A Machine Learning and AI Guide for Enterprises in the Cloud. Retrieved from TechTarget: https://searchenterpriseai.techtarget.com/definition/AI-Artificial-Intelligence

Sallomi, P. (2019). Artificial Intelligence Goes Mainstream. Retrieved from Deloitte: https://www2.deloitte.com/us/en/pages/technology-media-and-telecommunications/articles/artificial-intelligence-disruption.html

Simester, D. (2016). The Lost Art of Thinking in Large Organizations. MIT Sloan Management Review, 95-96 Summer.

Techopedia – Definitions. (n.d.). Retrieved from Techopedia: https://www.techopedia.com/definition/32099/automation

The Race For AI: Here Are The Tech Giants Rushing To Snap Up Artificial Intelligence Startups. (2019, September 17). Retrieved from CB Insights Research Briefs: https://www.cbinsights.com/research/top-acquirers-ai-startups-ma-timeline/

The Seven Patterns of AI. (2019, April 4). Retrieved from Cognilytica: https://www.cognilytica.com/2019/04/04/the-seven-patterns-of-ai/

Virtual Reality vs Augmented Reality. (n.d.). Retrieved from Augment: https://www.augment.com/blog/virtual-reality-vs-augmented-reality/

Dr. Thomas Collins was most recently Executive Director at HCC, a legislatively created information technology center, that provides IT and related support services to school districts, private schools, local governments, higher ed, and others in and around Cincinnati and southwest Ohio. Prior to HCC he was a consultant with Hamilton County ESC in Cincinnati, OH. He can be reached at (513) 967-5966 and collins2444@gmail.com.

Dr. Susan Leddick is a consultant in organization design and continuous improvement and President of Profound Knowledge Resources, Inc., a consulting and training firm. She has extensive experience working with educational service agencies, state departments of education, and other organizations. She can be reached at (512) 431-2879 and susan@pkrnet.com.

Leave a Reply

Your email address will not be published. Required fields are marked *