The AI Bill of Rights – all you need to know

In what could be termed yet another initiative to promote the use of responsible and trustworthy artificial intelligence (AI), the Biden administration has unveiled a new “AI Bill of Rights” which identifies some key principles to protect the rights of the American public in the age of AI.

The document, while acknowledging the great transformative power of artificial intelligence, provides a blueprint to avert potential harms caused by unaccountable algorithms and to protect civil rights, civil liberties, and privacy.

On Oct. 4 the White House’s Office of Science and Technology Policy (OSTP) released a set of voluntary guidelines, the Blueprint for an “AI Bill of Rights”, for governments at all levels to companies of all sizes. It identifies five principles to guide the design, use, and deployment of automated systems to protect all people against potential harms.

“Where existing law or policy—such as sector-specific privacy laws and oversight requirements—do not already provide guidance, the Blueprint for an AI Bill of Rights should be used to inform policy decisions,” reads an explainer by the OSTP.

The principles

The five aspirational and interrelated, yet non-binding, principles are: 1). Safe and Effective Systems 2). Algorithmic Discrimination Protections 3). Data Privacy 4). Notice and Explanation and 5). Human Alternatives, Consideration, and Fallback.  

According to the first principle, individuals should be protected from unsafe or ineffective systems. The second principle is about discrimination individuals could face by algorithms and stresses that systems should be used and designed in an equitable way. As per the third principle, individuals should be protected from abusive data practices via built-in protections and they should have agency over how data about them is used. The fourth principle stresses that individuals should know that an automated system is being used and understand how and why it contributes to outcomes that impact them. The fifth principle says that individuals should be able to opt out, where appropriate, and have access to a person who can quickly consider and remedy problems they encounter.

The concerns

While the White House’s new initiative drew widespread media attention it wasn’t without criticism.

For example, the WIRED termed the “AI Bill of Rights” as toothless against big tech. The Wall Street Journal quoting some tech executives wrote that “the nonbinding guidelines could lead to stifling regulation concerning artificial intelligence.”

“The Blueprint for an AI Bill of Rights notably does not set out specific enforcement actions, but instead is intended as a White House call to action for the U.S.,” reads a news report by the Associated Press.

As per the document, the blueprint was developed through extensive consultation—with stakeholders from impacted communities to experts and practitioners as well as policymakers—on the issue of algorithmic and data-driven harms and potential remedies.

GAO report on challenges and benefits of ML technologies in medical diagnostics

GAO report on challenges and benefits of ML technologies in medical diagnostics

Government Accountability Office (GAO’s) latest report has identified several challenges affecting the development and adoption of machine learning (ML) technologies in medical diagnostics. The report also offers some policy options for addressing these challenges and enhancing benefits.

The 103-page report “Technology Assessment Artificial Intelligence in Health Care: Benefits and Challenges of Machine Learning Technologies for Medical Diagnostics” to congressional requesters has been jointly published by the GAO and the National Academy of Medicine (NAM).

Cost of diagnostic errors

Emphasizing the importance of effective, efficient, and accurate clinical diagnostic process, the GAO’s report quotes a report by the Society to Improve Diagnosis in Medicine according to which each year diagnostic errors affect more than 12 million Americans with aggregate costs likely in excess of $100 billion.

The report says that challenges to the development and use of ML medical diagnostic technologies raise technological, economic, and regulatory questions. “… AI tools developed using historical data could unintentionally perpetuate biases, reduce safety and effectiveness for different groups of patients, and produce disparities in treatment,” notes the GAO report.

The challenges

Some of the several challenges affecting the development and adoption of ML in medical diagnostics found by the GAO are:

  1. Demonstrating real-world performance across diverse clinical settings and in rigorous studies as, according to experts and stakeholders, medical providers may be reluctant to adopt ML technologies until its real-world performance has been adequately demonstrated in relevant and diverse clinical settings.
  2. Meeting clinical needs, such as developing technologies that integrate into clinical workflows as medical providers are less likely to adopt ML technologies that do not address a clear clinical need, and many ML diagnostic technologies do not progress from development to adoption for this reason
  3. Addressing regulatory gaps, such as providing clear guidance for the development of adaptive algorithms as gaps in the regulatory framework may also pose a challenge to the development and adoption of ML technologies.

The policy options

As per the report these and many other challenges affect various stakeholders including technology developers, medical providers, and patients, and may slow the development and adoption of these technologies.

To address these implications, the GAO report has identified three policy options to help address challenges or enhance benefits of ML diagnostic technologies. These options include encouraging evaluation of these technologies, improving high-quality data access, and promoting collaboration across stakeholders.

  1. Evaluation: Policymakers could create incentives, guidance, or policies to encourage or require the evaluation of ML diagnostic technologies across a range of deployment conditions and demographics representative of the intended use.
  2. Data Access: Policymakers could develop or expand access to high-quality medical data to develop and test ML medical diagnostic technologies.
  3. Collaboration: Policymakers could promote collaboration between developers, providers, and regulators in the development and adoption of ML diagnostic technologies.
U.S. to leverage AI/ML to predict Ukrain’s weapons and ammo needs

U.S. to leverage AI/ML to predict Ukrain’s weapons and ammo needs

From smart systems to smart weapons, artificial intelligence (AI) is already impacting our lives in different unimaginable ways and this also includes the way how wars are planned and fought today.

With its great transformative power, AI and machine learning (ML) technologies have the potential to push nations to gain a strong position of advantage over the enemies, adversaries and peer competitors on and off the battlefield.

Besides fueling autonomous weapons to stealth drones and including a long list of matériel, AI and ML technologies are behind most of the modern day warfare systems while much is also being explored, almost everyday, to change the face and the pace of warfare.

Something similar is happening at the International Donor Coordination Center, or IDCC, set up at the US military barracks in the German city of Stuttgart where military personnel from different nations including U.S. Germany, Britain and France and others work together to ensure Ukraine gets what it needs to help defend its sovereignty.

As the Russian invasion of Ukraine enters over seven months now, the United States and its allies have been providing the defending nation with weapons, ammunition and other aid it needs to defend itself against Russia. It is at the IDCC where the U.S. officials are now working on AI/ML to predict Ukraine’s weapons and ammo needs.

“That’s partially why the next step is take the large amounts of data that the coalition is collecting and develop AI-driven techniques to anticipate those needs ahead of time, rather than just respond as they come in”, said Jared Summers, the chief technology officer of the 18th Airborne Corps, who is working with the IDCC in Germany.

The IDCC is the focal point of the entire multinational operation where it tracks the needs and delivery of weapons, ammo and other aid extended to Ukraine. In order to make the whole process more efficient and transparent and to predict Ukraine’s weapons and ammo needs, efforts are being made to AI and ML for this purpose. The idea is to translate valuable data into competitive advantage to save precious time and resources.

The invasion continues and so is the support from the allies. Besides training in how to use the weapons and ammo, Western allies at IDCC have helped deliver over $8bn worth of weapons and other aid to Ukraine’s armed forces. These include, but not limited to, thousands of autom

American AI Bringing new focus to privacy enhancing technologies

Bringing new focus to privacy enhancing technologies (PETs)

With technology becoming ubiquitous, the digital world landscape is fast evolving. Amid an ever-expanding data universe, a continuously available digital world has many new opportunities to offer. But there are challenges too. One of these challenges is our changing perspective on privacy.

Therefore, today there are more laws, than in the past, mandating stronger data and privacy protections. According to the U.S. Government Accountability Office (GAO) with the advent of new technologies and the proliferation of personal information, ‘the protection of personal privacy has become a more significant issue in recent years.’

In order to address such issues, the development and adoption of privacy-enhancing technologies (PETs) is becoming increasingly common. These technologies have the potential to safeguard data used in artificial intelligence (AI) and machine learning (ML) applications.

Defining PETs techniques

With the help of privacy-enhancing technologies, insights from sensitive data are gained while protecting privacy and proprietary information of individuals. These emerging technologies help data stay private throughout its lifecycle.

PETs also include maturing technologies, such as federated learning, which allow machine learning (ML) models to be trained on high quality datasets collaboratively among organizations, without the data leaving safe and secure environments.

As per the Internet Association of Privacy Professionals (IAPP), PETs refer to the use of technology to help achieve compliance with data protection legislation.

According to this report of the Federal Reserve Bank of San Francisco, some of the types of PETs include anonymization, encryption, multi-party computation and differential privacy.

Development of a national strategy 

Some recent initiatives including collaboration between the U.S. and the U.K. Governments on prize challenges to accelerate development and adoption of PETs and the White House Office of Science and Technology Policy (OSTP) issuing a request for information (RFI), last month, for public comments to help inform development of a national strategy on advancing privacy-enhancing technologies have brought new focus to PETs. These moves are expected to attract widespread adoption of PETs in the future across various departments and agencies.

As per the RFI, this national strategy will put forth a vision for “responsibly harnessing privacy-preserving data sharing and analytics to benefit individuals and society. It will also propose actions from research investments to training and education initiatives, to the development of standards, policy, and regulations needed to achieve that vision.”

In one of its recent blogs, OSTP explained that the development of privacy-enhancing technologies can provide a pathway toward a future where researchers can analyze a broad and diverse swath of medical records without accessing anyone’s private data. It also noted that PETs can provide a pathway toward this future by leveraging data-driven technologies like artificial intelligence, while preserving privacy.


Analyzing Storytelling About the Ukrainian Conflict

Note: This article series in no way minimizes the Ukrainian suffering. It is simply an attempt to analyze the storytelling applied by various entities in the Ukrainian Conflict.

Humans have been telling war stories for millenniums. Many of those stories were about glorifying the victories and the victorious. Some detailed the brutality of warfare. Others described the inherent passions of war. Since the advent of the Internet, storytelling has become more democratized, more personal, and more instant. With multiple types of unstructured data being uploaded by hundreds of thousands of eyewitnesses of wars to millions of commentators, a new type of storytelling has emerged. The frames and themes of war provide elements whose complex interactions generate narratives that show and document the atrocities of war, human suffering, punitive actions, moral standing, changes in social and global order, and other dynamics of war.

In the theatre of today’s great power competition, the Ukraine conflict offers a first look into the storytelling related to new geopolitical dynamics: a hot conflict where major powers found themselves on the opposing ends. When combined with the deployment of artificial intelligence based narrative analysis, the Ukraine war offers significant insights into the evolving global narratives and how war stories will shape the geopolitics in the future.

This is a series of two parts of articles. The first focuses on major story themes coming out from countries. The second part of the article analyzes the narratives and positions taken by companies.

Theme 1: It’s about the history (Russia’s emotionless story that backfired)

The Russian storyline was based upon showing NATO and the Ukrainian government as the aggressors who were threatening the national security and interests of Russia and attacking ethnic Russians in Ukraine, and therefore Russia had no option but to attack Ukraine. A story that depicts lack of options is often a tragedy that invites sympathy, but its appeal must be emotional. The Russian story came across as too rational and lacked any emotional appeal for the global audiences. President Putin’s attempt to position NATO as a goliath failed terribly. If NATO weren’t turned into a goliath, Russia could not be viewed as David.

Despite the claims that Russian bots were superactive and that Twitter removed over 75,000 accounts during the war, the overall Russian story and communications were highly ineffective.

The text coming out of Kremlin, the images, and the visual storyline, were not designed to determinedly project Russia either as a decisive aggressor (goliath) or as a victim (David). It was somewhat a mixture of the two. Here was President Putin explaining why Russia went to war, but his explanation seemed like a lesson from a boring history class. Dwelling over a century old Ukrainian and Russian history, he easily lost global audiences. His speech was not designed to inspire his nation or to even make a strong case for the invasion, but instead seemed like an attempt to provide a justification for the invasion that was based upon some type of a historical account. In a world where emotions and drama dominate the narrative setting discourse, Putin was walking into a landmine. His grievance centric appeal did not have the emotional depth to resonate with the global audiences. Maria Zakharova, Russia’s spokesperson, became emotional about the fact that her side of the story was not being acknowledged and the plight of ethnic Russian in Ukraine was being ignored. But by that time, it was already too late for global audiences to develop any sympathy for the Russia’s position.

On the visual storytelling side, President Putin’s posture of leaning back, even slouching, while speaking with hands holding the edge of the table depicted someone drowning who needed to hold on to something. It brought to mind the iconic scene from the movie Titanic, where Leonardo DiCaprio was holding on to the edge of the wooden panel, right before he let it go.  But by the time that scene takes place in Titanic, audience has already developed deep sympathy for DiCaprio’s role. With the established image of a strongman, President Putin carried no such sympathy.

Further, on the visual storytelling side, the shots for the speech switched between showing President Putin slouching back and looking tiny behind the desk, overpowered by a TV screen, computer screen, keyboard, mouse, and four phones. The noise in the imagery was intense. If the idea was to show President Putin in action, that was not the right time or avenue. If the goal was to show him deliver a strong message, the visual noise was intense.

At the ceremony of signing DPR and LPR, the visual storytelling seemed cold and impersonal. The size of the room presented the image of a lonely emperor conducting business in a giant palace. There were no cheers or human emotions. From a Russian perspective, two states were being born, but the presentation seemed so solemn as if a death sentence was being awarded. Contrast that with when American presidents sign bills, applauding supporters of the bills are shown standing behind the president. Similarly, the audience at bills signing ceremonies is composed of citizens who cheer and applaud.

Russia’s failure to make the alleged NATO push and the Ukrainian neo-Nazism an emotional sell to the world was obvious. Unlike the US president who talked to the global audiences, President Putin limited his messaging to his domestic audience. There was no emotional depth, no drama, and no framing in President Putin’s claim. His speeches looked like long and boring history lessons rather than the great oratory to inspire nations or to make a strong case.

If this was all done intentionally, it is hard to imagine what it accomplished. If it happened because of incompetence, it would probably be seen as a bigger failure for Russia than the poorly executed war itself.

Theme 2: “We Understand” (Minimizing the Ordeal, China)

The Russian story was communicated in China very differently than in many other parts of the world. The word “invasion” was not used in China, and the images shown on national TV were of Russian soldiers distributing food and water to the Ukrainians. On social media any criticism of Russian aggression was removed. The official line of the Chinese government continued to be: We don’t want war in Europe but also don’t want any sanctions on Russia.

This narrative is designed to position NATO as the aggressor and Russia as the optionless victim. The official line of the Chinese government attempted to position NATO as hegemonic and intrusive, and that NATO was the threatening and destabilizing force against Russia. This was architected to create a preemptive story about Pacific QUAD (not a formal alliance, officially the Quadrilateral Security Dialogue, is a group of four countries: the United States, Australia, India, and Japan) as a hegemonic and destabilizing force. This narrative was being framed by not only what was being said or communicated but also by what was not being said and shown. Thus, not communicating a story is itself a story.

The dominant narrative in China was shaped by silence and gentle nudge to request both parties to reach peace.

Theme 3: We have come to honor that allegiance (Indian business-ism)

In the second Lord of the Rings movie, Haldir says: I bring word from Lord Elrond of Rivendell. An Alliance once existed between Elves and Men. Long ago we fought and died together. We come to honor that allegiance. The messaging coming out of India was similar. As the war progressed, India increased its trade with Russia. It was more than business as usual. The White House’s reaction to India’s position was that they found it to be “unsatisfactory” but “unsurprising”. While it was hard for India to justify its position with its Western partners, the Indian story needed to be of pragmatism and opportunism with an undertone of amorality of political affiliations. Terminating its longstanding relationship with Russia was not an option. Neither was becoming a hard critic of President Putin. If you can’t take the Western side, might as well benefit from the situation – turned out to be India’s story. India was willing to trade its “Gandhi” image for the image of India is open for business. The Indian story was of pragmatism and astuteness. It was saying: you don’t have to trust us that we will always stand for the oppressed, but you can trust us that we will make good business decisions.

Theme 4: Please let us help you (Caught in the middle, Israel and Turkey)

Israel was caught in the middle. On one hand Israel needed to stand with America and its Western allies and on the other hand Israel had to maintain good relations with Russia. The Israeli story could not have been about taking sides. Unlike India which took the Russian side, the Israeli story developed as of a mediator and of a refugee host. In the mediator role, Israel offered to mediate peace between the two warring parties. On the refugee host side, the country’s situation gets even more complicated as about two-third of the Ukrainian refugees arriving in Israel are non-Jewish. Israel was able to keep a balanced position. Turkey’s position, like Israel’s, also turned out to be of a mediator.

Theme 5: We are against the war but stand with Russia (the anti-imperialist story, Iran)

Iran took a position of blaming the NATO for pushing Russia into a corner but also claimed that it is against the war and human suffering. This position allowed Iran to develop a narrative of standing for the oppressed without disparaging its close ally Russia. This was in line with the Islamic Republic’s overall narrative of anti-imperialism.

Theme 6: We are all victims of hegemonic powers (Pakistan’s story)

Pakistan’s prime minister Imran Khan used a unique angle and turned his story into a narrative of victimhood of weak nations suffering the consequences of a war between the two giants (America and Russia). It brought to mind the images from the first Hobbit movie where two mountain giants are fighting as Bilbo’s party tries to save itself from the falling rocks. Pakistan’s story of strategic neutrality with victimhood was designed to deflect the decision to choose. That positioning of a victim was also meant for the domestic audiences who are greatly impacted by the rising inflation in Pakistan. PM Khan is fighting for his prime minister position against a no-confidence motion in the parliament, and the broader global conflict allows him to blame the rising inflation on geopolitical realities.

Theme 7: I am on nobody’s side, because nobody is on my side, little orc (Hungary)

Just as Treebeard in the movie The Two Towers claims that he is on nobody’s side, Viktor Orban, Prime Minister of Hungary said “Russia looks at Russian interests, while Ukraine looks at Ukrainian interests. Neither the United States, nor Brussels would think with Hungarians’ mind and feel with Hungarians’ hearts. We must stand up for our own interests.” The story of Hungary is about focusing on its own interests. This is different from India’s story as it is based upon a clear and overt claim of self-interest whereas the Indian story is based upon preserving old friendships.

Theme 8: The David vs. Goliath (Perfect execution by Ukraine)

On the other side, Ukraine offered the story of Russian aggression and backed it up with strong emotional data. President Zelenskyy’s used the right messaging and imagery – both in text and the visual elements. President Zelenskyy ditched suit and put on a military t-shirt and allowed his beard to grow. The visual imagery of a leader fighting aggression was delivered perfectly. The Russian aggression was captured in video and images, in news and social media, and the story was backed by proofs and strong emotional content. It quickly became the most touching story in the world. Even those countries that did not vote against Russia at the United Nations condemned the Russian war against the Ukrainian civilians. For example Iran, a strong ally of Russia, offered help to Ukrainians. The power of the Ukrainian storytelling was amazing. It touched hearts and it appealed to reason. The Russian story was crushed by the power of the Ukrainian story. The Ukrainian story gained enough momentum for the US and EU to enact sanctions and terminate business with Russia.

But David vs. Goliath story requires David to have a stone and a sling, and for him to use that to hit Goliath with some force. For Ukraine to pull that story, a significant victory over Russia, even if turns out to be of resistance and defiance, will be critical. So far, the storyline of the Kiev defense is working out in Ukraine’s favor. Russia understands the risk and has decided to refocus the campaign in the eastern Ukraine and has announced that the Phase One of the war is over. The end of Phase One was not turned into a victory lap by Kiev. This could be because the timing of the story is slipping. The story has climaxed, and short attention spans of modern audiences quickly lead to cognitive saturation. The long-drawn fight and the words like “stalemate” and “stymied”, if not linked to a victory lap, may work against the Ukrainian story. For global audiences these words signal “move on to something else”. Suddenly, Will Smith punching Chris Rock in Oscars will become a far bigger story.

Important Considerations

The question here is that whether Ukrainian story can sustain its power? Emotional stories can lose momentum quickly as human emotions are designed to be reactive, but they can’t maintain a state of hyperactivity for too long – especially when things are not too personal. Slowly, people will move on to other stories. Alternative explanations will emerge. Counternarratives will rise. NATO will be blamed for fueling and extending the war. The stories about the economic toll of the war for Americans will start taking center stage. Would Americans be okay with paying nearly twice at the gas pumps in the first summer after two years of struggling with Covid related restrictions? President Biden’s approval rating has already fallen to the lowest level in his presidency.

Russia will undoubtedly try to take advantage of these conditions and rearticulate its story to a narrative that shows that the world’s largest democracy (India), the world’s largest economy (China), and a country that understands human rights better than anyone (South Africa) stand on its side. But that depends upon Russia recognizing the importance of storytelling. Since President Putin likes to talk about history, if the recent history is any indicator, Russian storytelling was a dismal failure.

American Institute of Artificial Intelligence (AIAI) is an institute focused on using Machine Learning to analyze stories and narratives of companies, countries, and government agencies.


Why analysts overlook the greatest opportunities?

Reindustrialization is enabled by disruptive innovation, and historically, great transformation times offer opportunities to create powerful returns for investors. It is often hard for analysts to develop a perspective that allows them to make sense of such powerful developments.

Why Financial Analysts miss the Reindustrialization Opportunity?

Analyzing a new technology is never easy but the current processes, approaches, and methods of analysis make it impossible for traditional analysts to understand the powerful dynamics of reindustrialization. Reindustrialization is not like any other times. It is a great transformation with a massive potential to create change. Here are some of the attributes that make reindustrialization analysis hard for traditional analysts:

Nature of Innovation: The nature of innovation in reindustrialization is not the same as in ordinary times. Analysts are used to analyzing innovations that do not exhibit the unique features of innovations that initiate and propel reindustrialization. At the most fundamental level, in reindustrialization the scientific process itself becomes more efficient or is enhanced by a more efficient process. This change happens at epistemological, ontological, and ethical constructs of science. The scientific efficiency enables new and unexpected technologies to appear on the horizon. The rate at which technologies appear and their novelty increase significantly. Analysts are trained to observe the technologies based on their adoption trends, market share, and features and function – and not on their scientific paradigms.

Analyst’s Positioning: Analysts occupy a certain placement in their corporate positions. Their respective position determines their vantage point. The reindustrialization dynamics are different as they requires simultaneously analyzing multiple industries, various sectors, multidisciplinary developments, and competitive structures.

Competitive Dynamics: The competitive dynamics of reindustrialization are different than normal competitive times. Specifically, the traditional competitive boundaries and moats do not exist during reindustrialization times. The shape, structure, and boundaries of industries and sectors are in flux. Nontraditional actors from one industry can enter other industries unexpectedly. Industry analysts are trained to think in terms of industries. As recent experience with Tesla shows, Tesla cannot be analyzed as an auto company.

Technical Characteristics: The technical characteristics of reindustrialization companies and innovations are widely different than that of ordinary innovations. For example, the artificial intelligence technology is about systems that are developed from data vs. regular IT systems that are made for data.

Operating Models: The operating models for reindustrialization are different. In many cases the innovation transforms the operating models itself. The operational and execution plans and strategies need to be analyzed from the reindustrialization viewpoint. Both product and production platforms change.

Go to Market Strategy: The market entry strategy of a firm, its positioning, and go to market plans change. The change happens at the most fundamental level.

Geopolitical Rivalries: Reindustrialization often resets the global friendships and rivalries and changes the rules of competition. Supply chains are realigned and remapped. Analysts often attribute such changes to political developments – even though such changes are often due to realignment of global competitive forces due to reindustrialization. For example, China and US relations changed with the advent of artificial intelligence and quantum computing.

Value Mapping: Understanding and mapping value creation in reindustrialization is not as straightforward as analysts are used to. It requires rethinking the production, distribution, and sales process.

Narratives: Reindustrialization carries its own narratives. Understanding and acting upon those narratives require developing new research methods. They include the use of natural language processing and ethnographic studies.

Behaviors (expectations): Deciphering shareholder, investors, and customer expectations is an important element of sensemaking.

Financial Models: The standard valuation models may not offer the best base to understand asset prices. Reindustrialization requires the introduction of new methods to study asset values.

The above factors greatly contribute to the inability of research analysts to understand the dynamics of reindustrialization.


How to put “S” in ESG? A guide for investment managers: AI does it for you

By Prof. Al Naqvi

Reindustrialization and ESG do not have to be incompatible. AI can enable you to accomplish both. This article only focuses on the “Social” or S part of the ESG.

The ESG movement has a strong emphasis on environment. ESG stands for Environment, Social, and (corporate) Governance.

This whole area is so new that when CNBC interviewed Jonathan Bailey, Head of ESG Investment at Neuberger Berman, the CNBC analyst commented that Bailey’s position would not have existed a year ago. (interview aired on 08/24/2020)

Clearly, getting the E part is easier. It has 40 years of advocacy by people like Al Gore, well-defined global standards, and tradable products such as Carbon Credits.

But what about S in ESG? In an era marked with Black Lives Matter protests, rising nationalism across the world, increasing global tensions, and higher awareness about these issues – how to bring S into ESG without feeling the guilt of insincerity?

“Guilt of insincerity” … what’s that?
Claiming to start an ESG focused fund is easy. Defining the standards in accordance with which the fund will operate is much harder. What makes it extremely difficult is that you must deliver reasonable returns to the shareholders.

To posit that great returns will only come from companies with the highest ESG performance is still unproven. To claim that firms with higher standards deliver great value, is also empirically unverified.

But what is provable is that if a strong movement exists in the investment world – a movement that can help transition investment from assets that exhibit low ethical standards to those that show strong ESG traits, the needle will move. The obvious question is: what are the standards for “S” – social concerns?

Without having such standards – fund managers are trapped in a debate that precedes Al Gore’s save the planet campaign by a few thousand years. The question of what is good for society is as old as the human civilization.

If ESG fund managers pay lip service to social concerns, they are being insincere to the cause. If they don’t do anything – then it is not much of ESG. If they create goals that are too tight, it may reduce the breathing room for delivering returns. This is the dilemma and a source of guilt.

Without more elaborate frameworks, asset managers tend to lean upon some obvious areas such as diversity, human rights, consumer protection, and animal welfare.
The United Nations provides further guidance on the sustainability development agenda by establishing 17 goals. But these are cookie cutter approaches. A fund needs a competitive advantage.

While these standards are clear, the following three problems become a source of concern for investment managers:

The problem of variable inclusion: What constitutes as social concern? Which social value driver variables to include and which to exclude? For example, would running clinical tests on animals be considered as a concern about animal welfare? Would running clinical trials in Africa – where populations may not really understand what they are getting into – constitute as human rights violations? Would capturing personal information of users be considered a human rights issue?

The problem of measurement: Once variables are defined – questions arise about how to measure the social impact? Would the measurement be against an absolute standard or would there be some flexibility?

The problem of definition: This is close to the first problem – but it captures a different perspective. While the first addresses which variables to include, this problem addresses that once a decision is made to include a variable, how do you define that variable. For example, if you have selected weapons manufacturing as a (negative) social value driver variable, does all weapons manufacturing violate human rights or only selling weapons to countries that violate human rights constitutes as bad? It is questions as that which help us define the variables.

The above three considerations impact the real problem faced by fund managers: delivering returns above the expected cost of capital.
This means that answers to the above should somehow be linked with returns – and that means answering various strategy and goals of fund questions. For example: Is the fund operating with the goal of behavior modification for a target firm – or is the goal to reprimand offenders? Does the firm establish internal goals or abide by external goals? And most fundamentally: how does the ESG-enabled strategy translate into a competitive advantage?

The answer to the above question lies with artificial intelligence.


How can AI help?
Artificial Intelligence provides the most comprehensive way to implement “S” and “G”. In this article I will only address S – as G will be discussed in a separate article.
I believe that every firm needs its own internal standards and the ability to analyze target investments. This gives maximum strategic flexibility and can help firms establish standards that can be unique.

A system for achieving that is based upon 5 Social Value Discovery drivers – which are followed by a CRISP-DM type model development and deployment process (Figure 1).The first four steps are what helps define the overall social value creation model.
The fifth step – Return Linking – establishes a hot link between social value drivers and returns to enable dynamic fund management. The word dynamic refers to having the ability to evaluate the link between social value creation and returns. This strategy can work for both passive and active investment styles.

The outcome of the first five steps establishes the data requirements and scopes out the preprocessing requirements – keeping in mind that your data and algorithms can themselves be a source of bias.

You must avoid these mistakes

To secure solid returns and keep your “social” a strong contributor of value – you must avoid the following five mistakes:

Do not go with a cookie cutter approach. Your investors will be able to smell the insincerity associated with the cookie cutter approach. Make social value creation a competitive advantage for your firm. This means you must have a framework.

Do not leave the 5th Step as a loose end. Make sure to link returns with your social value drivers – and do so at a lower level in fundamentals-based value creation. Do not use indefensible models of social value creation. Make sure to have clear and defensible framework which clearly defines what value creation means.

Do not proceed without properly understanding the data and algorithms: Your data and algorithms can be a source of bias. You must have a clear strategy on how to deal with bias.

Make sure to test your strategies: Make sure that you have a way for testing your strategies. And this opens a new can of worms. Your model must work in active and live settings – and making it work is not easy. Good investment strategies often come to die on the altars of overfitting.

Do not ignore that your target investments can use AI for good and bad: One of your critical evaluations need to be whether your target investments (firms) are using AI to create or destroy social value. These days AI is the top agenda item for most firms – and they can use it for good or for bad – and your knowledge of that can make all the difference.

Next Steps

Based upon the above, here are the five steps to put a solid S in your ESG:
1. Understand your strategy; identify your social value drivers, measurements, and definitions.
2. Establish a hot link with returns. Study what that means. Test, test, and test. Do not use a cookie cutter approach.
3. Data is expensive. Do not just get all the data. Get data that is meaningful for your framework. Establish best practices for data management and preprocessing.
4. Establish models – and you would need multiple – to work together to identify and manage value creation for you. This means to get a synchronized value-creation framework implemented.
5. Things are never constant. Know and proactively manage when change happens. It can happen when the underlying distributions have changed – or the set of features you used to define the social context have been altered.
You mean well. But this is investment business and having goal excellence is one thing – making it work another. Social value creation is important and while we cannot change everything – as Dylan Thomas said, we must “rage against the dying of the light”



How to design ESG programs that don’t get your CEOs fired? Here is the clue: Use AI.


At least two major activist investors (Bluebell Capital Partners, Artisan Partners) joined the calls for the removal of Emmanuel Faber, CEO of Danone (the French firm known for its yogurt products and bottled water). The board heard the message loud and clear, and Mr. Faber was removed from his position (see Figure 1). During Mr. Faber’s tenure, Danone’s stock value increased 11% – meanwhile during the same period its competitors Nestle and Unilever increased by 43% and 55% respectively. Since underperformance often leads to the removal of CEOs, that is not what is special about this firing. What makes this unique is that Mr. Faber was the icon of stakeholder activism, sustainability leadership, and ESG management.

He is not alone. We have observed a recent trend of sustainability champion CEOs being toppled. For example, Isabelle Kocher of Engie and Sacha Romanovitch of Grant Thornton, both great supporters of the sustainability and ESG movement, were also ousted.

Does this signal that Milton Freidman is back and the ESG movement is dead?

I believe that is not the case.

Figure 1: Difference between March 1st and March 15th press releases (source

What are ESG programs missing?

After years of research on this problem, I have concluded that the real culprit is how ESG programs are approached. The standard designs of ESG programs fluctuate between “denial and avoidance” and “check-the-box compliance”. Neither brings out the creativity and power that sustainability movement gives to a firm.

The problem happens because ESG program designs have become unauthentic and removed from business realities. They are designed to be responsive to large and powerful rating agencies and standard setting bodies and not to the needs of the business. They are not integrated with the business strategy. They lack authenticity and creativity. They have turned into large checklists designed to cater to the never-ending demands built upon the limited visions of rating agencies. More vocal CEOs try to lead with strategic visions of sustainability, but their programs turn into either lofty dreams of grandeur, or boring checklists. In the former case organizations do not buy into the programs and resist them with full force. In the latter case no one is inspired to take ESG seriously.

The question is: how can you design ESG programs that work? There are 5 steps.

Step 1: Start by analyzing your value chain

Many firms start their program designs by scoping out materiality and understanding standards. This can lead to problem programs that get CEOs fired. Start with first analyzing your business’s value chain. Look for the opportunities where you can add sustainability value by fine-tuning your value chain.

Do not start with stakeholder materiality analysis. While knowing what is important for stakeholders is important, a program designed from one-time analysis of stakeholder priorities and preferences will have a structure but it will lack flexibility. Since both stakeholders and their values are constantly shifting – such a design will lack the much-needed adaptive capability. Conversely, a program designed to be fully responsive to the changes in stakeholder priorities and preferences will lack much needed structure.

Similarly, avoid starting your programs with the sole purpose of complying with the standards. The ESG field is experiencing a standards proliferation – as anywhere you turn there seems to be new ESG standards thrown your way – and hence the starting point could not be standards. Standards are meant to be guidelines until they are not. If they become too rigid, they will lack “generally applicable” characteristic. If they are too loose, they will lack any enforceability. Unlike the financial accounting standards that are based upon theoretical foundations of how to classify transactions and report on financial performance, ESG standards lack that rigor. They tend to be practice and process guidelines.

Just as you do not design your business strategy based upon how numbers are reported in financial statements, your starting point should not be standards. Analyzing your value chain gives you the first view of ESG priorities for your firm. This means you can integrate ESG into your overall business strategy.

Step 2: Create a strategy

Once your initial analysis is complete, create a broad strategic plan that ties in an integrated ESG-Value Chain vision. This integrated value creation system does not approach sustainability as separate from the core business strategy. Instead, it views the transformation as central and core to the business strategy. The plan must have clearly defined metrics that tie into financial value creation for the firm.

Step 3: Create a broad narrative

Understand and manage broad themes and create narratives around them. Narratives show strategic intent and give meaning to initiatives and numbers. Without narratives, reporting is meaningless. Narratives are also linked to the business results.

Step 4: Link with stakeholder and standards

Now you can link the program with stakeholder preferences and any applicable standards. Again, these links are not the drivers of your plan and strategy but only a result. Doing the right thing should not be dependent upon how various stakeholders would feel about it. It should be done anyway.

Step 5: Transform your firm to the modern economy

ESG is a key ingredient in building a modern firm. As you set your strategic transformation program – make sure that ESG is well integrated into the DNA of your firm.

Using AI for ESG Management

The above 5-steps cannot be achieved without relying upon data and using the right tools. Fortunately, the advances in machine learning have given us the ability to design and manage successful and integrated ESG programs. AI/ML helps in implementing the five steps and ensures rapid and deep stay power of the ESG program.

After all, sustaining sustainability program should not get CEOs fired. The key is to identify the integrated value.


Nonergodic, history retaining trajectories of Disruptive Innovation

By: Al (Ali) Naqvi

My goal in this article is to expand the traditional disruptive innovation investment analytical framework with the additional constructs of nonergodic nature of disruptive innovations. Funds have a tendency to formulate investment thesis based upon a technology’s innovation potential and not necessarily its innovation path. Potential based analyses absorb all the nuances of behavioral elements, hype, marketing, and promoting a technology. Its reasoning mechanism is often comparative in terms of using a baseline historical event to establish performance potential of the disruptive innovation. When viewed from a path perspective, the analysis becomes far more complex – but the benefit of path-based analysis can result in far greater investment value.

The potential based analysis

Artificial intelligence has often been compared to electricity. Just as what the electric current flowing through the wires did, AI can revolutionize everything else. Some scholars and practitioners term such innovations as General Purpose Technologies, transformative technologies, or innovation platforms. For example, Brett Winton of ARK Investments argues that in the course of the last two centuries such innovations have triggered major market capitalizations (Winton, 2019). And today, he points out, waves of several transformative technologies are cresting, leading to a massive potential of market capitalization.

ARK identifies the attributes of disruptive innovations as being:

  • Across multiple sectors,
  • Upend existing or incumbent providers,
  • Create new business potential,
  • Deliver dramatic cost reductions,
  • Serve as a platform for new innovation, and
  • Propel global economic growth.

ARK keenly observes that dramatic decline in prices of such technologies creates rapid adoption, making existing technologies obsolete while creatively destroying established competitors. Similar observations were also made by Carlota Perez when describing the dynamics of technological revolutions  (Perez, 2002).

Fig 1 Adopted from ARK Investments (Winton, 2019)


Fig 2 Adopted from ARK Investments (Winton, 2019)

The above investment hypothesis, while impressive, covers only half the story. What it fails to capture are the distinct dynamics of the artificial intelligence revolution. These dynamics are not only widely different than anything we have experienced before, but are also far more complex than the subtleties and developments of technologies with ubiquitous adoption leading to market structures that gave rise to traditional public utilities.

The equilibrium is transitionary

In this regard the term “history matters” is more applicable than “history repeats itself”. Perhaps, history repeats itself in the sense that history does not repeat itself. Yet, history plays an important role, but not in the comparative sense of what transpired in the prior innovations somehow applies to the current innovation under the microscope, but instead in the sense that the evolutionary and dynamical development of the current disruptive innovation greatly depends upon its own history. The history preserved in the trajectory representing the evolutionary dynamics greatly affects the development and potential of the disruptive innovation. The transitions to the next states of development, adoption, and growth depends upon the prior states and the influences and conditions in the current states.

Unlike the stable equilibrium often claimed by the analysts, the development of technology follows more dynamical and evolutionary equilibrium where equilibrium’s stability is defined by the change itself.

Thus, discovering the right investment opportunities is not as much a function of generalizing broad equilibriums, as it is trying to understand the next transitory state and modeling the history and current state accurately to identify the specific trajectory of the innovation. Identifying this nonergodic development path is where the pool of opportunities is discovered.

Disruptive innovation is path dependent

A dynamical process whose evolution is governed by its own history is ‘‘path dependent.’’ (David, 2007). We have been cautioned against mixing economic history and economic theory. Nathan Rosenberg illuminated that innovation is path dependent and that the nonergodic nature of technological change requires analytical models to capture the inherent uniqueness and complexity (Rosenberg, 1994). Thus, analyzing the status of innovation at each state of its transitory movement to the next state can help clarify the trajectory. History matters. Furthermore, at each state of the path the numerous forces and state interval specific factors contribute to setting the subsequent direction of the innovation. Such forces may include previously committed investment, institutions, initial conditions, noise, and many other factors.

The scientific process is changing

The traditional scientific process is driven by seeking data as a function of hypothesis and experiment design. In the new era, scientific discovery can materialize from large datasets such that insights and findings precede hypothesis and experiment design. This reversal of the process introduces epistemological transformation and leads to hyper-acceleration and novelty. It can also create explain-ability challenge where innovators have to grapple with the findings that seem to hit the mark but lack theoretical explanation or justification.

Geopolitical constraints must not be ignored

Related to path dependent trajectories, geopolitical developments often alter the expected growth patterns. The emergence of China US trade tensions and eventual blocking of several Chinese firms by the US serves as a reminder that transitions in trajectories can introduce new patterns of unanticipated dynamics at the transition stages.

Noise impacts outcomes

Social systems are influenced by noise. Noise can result from faulty analysis and methods, misaligned incentives, managerial oversight, and other issues that can affect the transition into the next stage. Some of the path distractors can come from unanticipated forces. The highly admired dynamics of global brain to “help organize our social organism into a more coordinated, more efficient, more democratic, and more collectively potent entity” and in its ability “to foster more numerous and more diverse communications between both humans and technology, and then better link those communications to mechanisms of action”  – do not necessarily lead to stability (Rosenblum, 2015).  As seen in the recent retail investor mob raids to artificially raise the prices of assets – the global mind can become a counter force to stability.

Institutions matter

The nature of institutions plays a major role in disruptive innovations and technological growth (North, 1990). The governing philosophy, operational mode, and strategic outlook of institutions greatly influence how innovation transpires in the economy. Technologies affect institutional performance and institutions impact technological trajectories. In this bidirectional nudge pattern, the role of institutions to both embrace and influence innovation requires constant monitoring.

Enabling technologies and data

The states of development and adoption of support and complementary technologies that form the production platform of the primary disruptive innovation are critical analytical factors (Mowery and Rosenberg, 1989). Since many of those complementary and support technological innovation now depend upon the availability of datasets, the growth potential and adoption trajectories can produce different outcomes.


The initial conditions at the beginning states of the trajectory can influence the subsequent states and that is why it is important to take a note of the initial conditions. There were many advanced economies with highly educated workforce, but the performance of India in the early stages of the computer and internet revolution greatly impacted the subsequent history where India emerged as a powerful player in the software development industry.

Reindustrialization Dynamics are different

This leads us to the main point I am trying to make. Reindustrialization takes into consideration both history and the evolution of history within the time segment or states being analyzed. That is why, while the relationships and projections presented by ARK in Figure 1 and 2, can be accurate for a time segment, the number of exogenous and unknown variables impacting the transition state are enormous. In my next article I show how to measure and track the trajectories that enable phase transition.

But the same element that produces the ailment also gives the cure. Deep learning can help assess the state of a fast-changing reality, assuming a wide enough net is thrown to capture data. ARK’s (or other disruptive innovation investors) assertions, therefore, must constantly be tested, measured, reevaluated, and reported. This means analyzing and measuring innovation trajectories with a path dependence perspective. Path must integrate with potential to create investment value.



David, P. A. (2007) Path dependence : a foundational concept for historical social science. Cliometrica. [Online] 191–114.

Mowery, D. C. & Rosenberg, N. (1989) Technology and the pursuit of economic growth. Cambridge University Press.

North, D. (1990) Institutions, Institutional change, and Economic performance. Cambridge University Press.

Perez, C. (2002) Technological Revolutions and Financial Capital: The dynamics of bubbles and golden ages. Northampton, MA, USA: Edward Elgar.

Rosenberg, N. (1994) Exploring the black box: Technology, economics, and history. Cambridge University Press.

Rosenblum, F. (2015) Power and politics: A threat to the Global Brain. Technological Forecasting and Social Change. [Online] 11443–47. [online]. Available from:

Winton, B. (2019) Disruptive Innovation: Why Now? [online]. Available from: [online]. Available from:


Neuralizing a Private Equity Firm

By Al Naqvi

This article reviews how to reindustrialize a private equity firm?


Information technology (IT) is no longer a source of competitive advantage for companies. Too easy for competitors to copy. Too clunky. Too limited. Unlike artificial intelligence (AI), the legacy IT is deterministic and incapable of learning, adapting, or accumulating experience.

Neuralizing a company is the science of inducing behavioral modernity in the performance of a firm by using intelligent machines. It happens when a firm evolves to deploy integrated artificial intelligence infrastructure to automate and enhance work. This unleashes a powerful new wave of value creation.

Neuralization impacts six different aspects of human work (Figure 1):

Physical and Cognitive Automation: Automates existing work processes where work is composed of physical and cognitive components.

New Process Enablement: Enables new processes.

New Business Models: Allows firms to develop new business models.

Make New Scientific Discoveries: Accelerates new scientific discoveries.

Expands Human Cognitive Capacity and Awareness: Empowers executives with higher cognitive capacity and situational awareness.

Helps Improve ESG Value: Enables firms to develop stronger ethical and governance frameworks, enhance the effectiveness of corporate social responsibility initiatives, and improve positive environmental impact.

To read the full article please download pdf here