The Future of Intelligence Analysis
Anamaria Berea, PhD
Center for Complexity in Business, University of Maryland
3450 Van Munching Hall,
College Park, MD, 20742
Ph: 571-314-3768
Email:
ABSTRACT
While exploring various future scenarios can be an exercise in imagination, this research relies on hard evidence and quantitative data about the likelihoods of future technologies and future risks and challenges in the world as they are being forecasted by the TechCast platform, the oldest expert-based forecasting system in the world, with an excellent track in predicting the arrival of certain technologies or the emergence of global threats. While I construct the scenario of the future based on TechCast expert data, I infer the skills of the analyst of the future that she would need to have in order to address these most likely scenarios.
INTRODUCTION
The intelligence officer of the future will have to navigate an increasingly complex world, where there will be new jobs, new industries and new technologies that we can only begin to imagine now. Unlike other previous times in history, the world is moving and changing at a faster pace than ever and many of the jobs of today did not even exist 5 years ago, while many jobs of tomorrow do not yet exist today (Mokyr, 2015; TechCast Global, 2016).
But one thing that has not changed and will probably not change is the ability of the intelligence analyst to be creative, adaptable and interdisciplinary in his or her approach to problems (Clark, 2012). The intelligence analyst has been - at the core - an interdisciplinary researcher or a creative agent since the beginning of time (Barger, 2004). But the way the analyst is addressing the threats and the system under which she operates has been changing faster than ever.In the 21st century, the intelligence officer will have to address new threats that will come from new industries and the continuous fragmentation of the states (Barger, 2004; TechCast Global, 2016).
At its core, the intelligence process is one of gathering information, connecting the dots and getting actionable plans and scenarios ahead (Clark, 2012). But, arguably, the most important skill of the intelligence offices is forecasting uncertainty (Fate & Jackson, 2005); at the same time, forecasting has shifted more towards making sense and getting more meaning out of lots of noise and data, with less emphasis on the actual gathering of data (Berea et al., 2012) and this trend is likely to continue in the near future.
Therefore, in order to understand the necessary skills of the intelligence analyst, we must understand how the future world looks like and we need to answer these questions:
- What will be the increasing trends of the future?
- What will be the increasing threats of the future?
- And how much of the status quo will be preserved?
- How can we connect the dots and separate the noise from the meaningful information that is available?
- What kind of information do we expect to be difficult to collec?
In other words, forecasting not only what is going to happen, but also forecasting what information would be available and what information would not is one of the greatest challenges that the future analyst will have to address. This means that one of the most important skills of the analyst would be the capability to answer one of the greatest questions in uncertainty analysis: “what do we know that we don’t know?” – the “known unknowns” (Boardman, 2005).
In this paper I am exploring the problem of the future skills of the intelligence officer twofold: 1)Based on big, integrated pictures of the world of tomorrow (forecasts of big trends, technologies, and organizations) and secondly, 2) Based on the sources of information and addressing not only what is known or unknown at a specific point in time, but forecasting the information availability itself - what will be known or unknown at a certain point in the future, what information will be private and how much information can be harvested and analyzed automatically versus human interpretation.
METHODOLOGY
TechCast Expert Forecasting System
In order to address specifically questions No. 1 and 2 about the trends and threats of tomorrow, I am using the TechCast forecasting system (TechCast Global, 2016)
TechCast Global is an expert-based forecasting platform that has been gathering forecasts regarding some of the most important trends in technology and society and their likelihood ofhappening in the future (Halal, 2013).
Lately, prediction markets have been used extensively to forecast short-term questions and very specific questions that would help any decision maker, not just intelligence officers, implement better actions and policies (Berea & Twardy, 2013). But TechCast has the advantage of forecasting very long-term questions, very big trends and also has the advantage of having done this for a very long time (Halal, 2013).
The TechCast platform also has quite a good record in forecasting accuracy. While the experts from TechCast are focused on bringing forth the latest in-depth analyses, they are forecasting based on arrival times of certain technologies and on how many years in the future something is going to happen or not. This forecasting accuracy falls on a mean of +/- 3 years off forecasts, with most forecasts falling within the +/- 2 years in range.
Quantum and Bayesian computing
In order to address the second point with respect to forecasting information availability or non-availability, privacy or transparency and automated or human, I am relying in my analysis on the future trends in Bayesian quantum computing and in artificial intelligence (Trueblood et al., 2016). There is undoubtedly an increase in the use of quantitative methods and automated methodologies, such as machine learning, that process and select information and even recommend the best course of action (Japkowicz & Stefanowski, 2016). But these models can only do as much as learning from the past and new evidence informs them to do (Pearl et al., 2016). Therefore, in order to address the biggest question of all – how to estimate what we won’t know – will require human experience and analysis in understanding both the trends and threats of the real world AND the trends and threats from information sources or lack of at the same time.
Metaphorically, we can compare the forecasting the information, the absence of information and the trends and threats that these point to with a cosmological model of matter – dark matter and the evidence and signals that is brought forth by both of them.
TRENDS AND THREATS
The strategic world of tomorrow
The TechCast experts estimate that the G-20 countries will continue to grow and reach an estimated 3% growth per year by 2019. Nevertheless, China is expected to slow down its growth, EU debt to increase and overall the world to start slowing down and decrease its growth after 2020. This means that the world might face increased inequality and a shift in research spending from government to private resources (see Figure 2).
Overall, this means that the analyst of tomorrow would have to be prepared for threats due to inequality and industrial/commercial espionage. With the increased move of science and technology innovation in the private domain and with the increase of income and lifestyle inequality between countries and within countries, the threats coming from the privatization of vital information and the upsets of lifestyles means that the analyst of tomorrow will have to be prepared to understand the dynamics of a world of fragmented information and fragmented incentives and lifestyles.
The entrepreneurial world of tomorrow – new organizational forms
TechCast experts also estimate that the world is becoming “older” – the aging population will reach 25% of the world population by 2022. At the same time, Generation X will become more influential in their endeavors and they are notorious as serial entrepreneurs. The Millennial, on another hand, see themselves as “global citizens” and this means that the boundaries between the countries, ethnicities and religions will become more and more blurred. This generational interplay and shift is more likely to give us a world where private companies that easily cross boundaries are more likely to engage in espionage and cyberthreats against other companies. On another hand, this might also give us a more open and transparent world where the people seeing themselves as “global citizens” are driving agents for coexistence and coevolution.
For the analyst of tomorrow, this means that industrial espionage and the understanding of the “known unknowns” when there is an abundance of information, but which is private, and the source of this information (private – public; institutionalized – crowd-based) will come into first stages of their training.
There is also an increase in the “crowds” and not necessarily in “organizations” or institutions (Jackson, 2014). From this perspective, the analyst of tomorrow will have to understand the dynamic of the crowds and crowd behavior, which is different from the one of institutions. Crowd behavior is fast, transient, volatile and extremely powerful, and thus the analyst will have to receive training into understanding and modeling the sociology and dynamics of the crowds.
The beliefs networks of tomorrow – socio-ecological networks
The forecasts for extremisms growth after year 2020 are small. Due to migration between countries (20% of world population will leave elsewhere than they were born by 2026), increase in travel (to $10 trillion$ by 2021), urbanization (60% worldwide by 2022), education and virtual education increase (30% by year 2021) and generational shift, the world of tomorrow is more likely to become an integrated, inclusive world of beliefs. There is an expected increase in global ethics by 30% by the year 2033 and an increase in social responsibility by 50% by the year 2020.
More than just from extremism, an important threat will come from the financial and ecological fragile systems. The experts estimate that the energy, transportation and financial systems are the ones that have most criticality with respect to failure. They are forecasted to fail by on a rate of 33% within the nest 5 years. This represents one of the biggest threats the analysts of tomorrow will have to understand and avert.
The wars and conflicts of tomorrow
The experts estimate a 58% probability for the mankind to develop energy weapons, with an expected impact that amounts to almost half of the impact of the nuclear weapons. Energy weapons are electric lasers or laser weapons, that can be both lethal and non-lethal. Therefore the analyst of tomorrow will have to understand the potential of these weapons, the sources and targets for these weapons in a similar fashion with the weapons of mass destruction that we understand today.
Another cause of war that is likely to appear within the next 8 years is water. The experts estimate that water wars. OECD and World Bank estimate that almost 4 billion people will live under shortage of water or water stress by year 2025 (OECD, 2008).
While energy and water wars are new potential threats that the future analysts will have to prepare for, some of the current threats such as cyber-wars and bio attacks will continue. Therefore the analysts of tomorrow will have to keep on understanding and improving the ways they fight these current and new threats.
The intelligence team of tomorrow
Traditionally, organizations have had a hard time being very adaptive to the changing times (Heckman et al., 2016). Nevertheless, the intelligence teams of tomorrow would have to adapt to a mix of crowds and organizations (Jackson, 2014), to new ways of working in teams and new ways of being transparent while working with the private information and privacy issues. TechCast experts estimate that there is going to be 50% less privacy by 2018 and 51% transparency in government and governance by 2022.
Some of the status-quo in intelligence analysis will have to be preserved: skills like “connecting the dots”, highly collaboratively teams, working collectively with other teams or with more transparency and open source information will have to not only be preserved, but also enhanced.
On another hand, new skills from understanding fast changing scenarios in social and ecological shifts and in strategic players will have to be acquired. Undoubtedly the analysts will also have better “tools” and better technologies to forecast the future (artificial intelligence, automated algorithms and next generation computing will likely affect all jobs (Hanson, 2016)), to assess alternative scenarios and to determine the best courses of action, the best integrative assessment will still be made by the analyst team and the cohesiveness and adaptability of the team.
In a nutshell, the analyst of tomorrow should work in small, adaptable teams, be skillful in working with smart technologies (such as quantum computing) and tools to assist her in her job, to understand the dynamics of crowd behavior, to understand the dynamics of inequality and of industrial espionage, to continue understanding the new possible conflicts and wars such as energy and water wars, to understand the fragility of social and ecological systems such as finance, transportation and climate disasters. But, most of all, to understand where the “blind sidedness” of information might come from, in a world that is full of noise, with privacy and transparency constraints and with a wealth of privately generated data.
CONCLUSIONS
As this analysis shows, one the greatest challenges would still remain to recreate the whole, global pictures from myriads and bits of pieces of information, while at the same time to be aware of the information that is not readily available and that can be only collected from the field or private sources. The increase in the private information due to the shift of strategic industries and innovations into the commercial and global space will mean that the analyst teams of tomorrow will have to be ready to deal with information that comes or hides in many shapes and forms. Whilemost information and data from open sources will remain readily available, creating automated tools to support the analysts to recreate complete pictures from these data and connecting fast incoming data might be one of the greatest challenges for the future intelligence analysis.
While finance, economics and information transcend borders with few barriers, and the new entrepreneurial organizations and crowds have greater fluidity on the global stage as well, it is not the same when it comes to country security and country border definition. Organizations and individual roles now transcend border more easily and faster than ever before. At the same time, new challenges posed by new technologies, such as the artificial intelligence developments, space commercialization, delocalization of jobs, new rules of law, new economic risks and new types of companies and businesses in the context of an increasingly integrated global world will spur a new type of intelligence analysts and analyst teams – agile, adaptable, with constant awareness of the informational gaps and, just as like many other jobs of the future, assisted by the smartest technologies available.
REFERENCES
***, OECD ENVIRONMENTAL OUTLOOK TO 2030, ISBN 978-92-64-04048-9
***, , retrieved April 14, 2016
Barger, D. G. (2004). It is time to transform, not reform, US intelligence. SAIS Review, 24(1), 23-31.
Berea, A., Maxwell, D., & Twardy, C. (2012, October). Improving Forecasting Accuracy Using Bayesian Network Decomposition in Prediction Markets. In 2012 AAAI Fall Symposium Series.
Berea, A., & Twardy, C. (2013, April). Automated Trading in Prediction Markets. In SBP (pp. 111-122).
Boardman, M. (2005). Known unknowns: The illusion of terrorism insurance. Georgetown Law Journal, 93(3).
Clark, R. M. (2012). Intelligence analysis: a target-centric approach. CQ press.
Halal, William. (2013). “Forecasting the Technology Revolution: Results and learnings from the TechCast Project”, Technological Forecasting & Social Change 80 (2013) 1635–1643
Heckmann, N., Steger, T., & Dowling, M. (2016). Organizational capacity for change, change experience, and change project performance. Journal of Business Research, 69(2), 777-784.
Hanson, R. (2012). The Age of Em - Work, Love and Life when Robots Rule the Earth, Oxford University Press
Jackson, M. O. (2014). Networks in the understanding of economic behaviors. The Journal of Economic Perspectives, 28(4), 3-22.
Japkowicz, N., & Stefanowski, J. (2016). A Machine Learning Perspective on Big Data Analysis. In Big Data Analysis: New Algorithms for a New Society (pp. 1-31). Springer International Publishing.
McFate, M., & Jackson, A. (2005). An organizational solution for DOD's cultural knowledge needs. Military Review, 85(4), 18.
Mokyr, J., Vickers, C., & Ziebarth, N. L. (2015). The history of technological anxiety and the future of economic growth: Is this time different?. The Journal of Economic Perspectives, 29(3), 31-50.
Pearl, J., Glymour, M., & Jewell, N. P. (2016). Causal Inference in Statistics: A Primer. John Wiley & Sons.
Trueblood, J. S., Mistry, P. K., & Pothos, E. M. (2016). A Quantum Bayes Net Approach to Causal Reasoning. In Contextuality from Quantum Physics to Psychology (pp. 449-464).
ANAMARIA BEREA, PhD
Anamaria Berea has a PhD in computational social science from George Mason University and a PhD in International Business and Economics from the Academy of Economic Studies in Romania. After graduation, she was part of the George Mason University team on 2 IARPA funded projects – ACE competition and ForeST. Her research includes agent-based models of company growth, information crowdsourcing, prediction markets, Bayesian networks, social network analysis, large scale ("big data") analysis, text and sentiment analysis, and framing qualitative into quantitative modeling, diffusion of fashion, social media impact on crowd-funding success and the emergence of language and communication in socio-biological networks. She is a Teradata University Network Faculty Award Winner and her work has been published in Journal of Washington Academy of science, Decision Analytics, AAAI Proceedings, Quantitative Finance, Handbook of Human Computation and Journal of Strategic Security. Her research has been supported by grants from ONR, IARPA, DARPA and the National Academies of Sciences. She is a member of the Washington Academy of Sciences and Eastern Economics Association.
1