Managing Bias in AI: Strategic Risk Management Strategy for Banks
AI is set to transform the banking industry, using vast amounts of data to build models that improve decision making, tailor services, and improve risk management. According to the EIU, this could generate value of more than $250 billion in the banking industry. But there is a downside, since ML models amplify some elements of model risk. And although many banks, particularly those operating in jurisdictions with stringent regulatory requirements, have validation frameworks and practices in place to assess and mitigate the risks associated with traditional models, these are often insufficient to deal with the risks associated with machine-learning models. The added risk brought on by the complexity of algorithmic models can be mitigated by making well-targeted modifications to existing validation frameworks.
Conscious of the problem, many banks are proceeding cautiously, restricting the use of ML models to low-risk applications, such as digital marketing. Their caution is understandable given the potential financial, reputational, and regulatory risks. Banks could, for example, find themselves in violation of anti discrimination laws, and incur significant fines—a concern that pushed one bank to ban its HR department from using a machine-learning resume screener. A better approach, however, and ultimately the only sustainable one if banks are to reap the full benefits of machine-learning models, is to enhance model-risk management.
Regulators have not issued specific instructions on how to do this. In the United States, they have stipulated that banks are responsible for ensuring that risks associated with machine-learning models are appropriately managed, while stating that existing regulatory guidelines, such as the Federal Reserve’s “Guidance on Model Risk Management” (SR11-7), are broad enough to serve as a guide. Enhancing model-risk management to address the risks of machine-learning models will require policy decisions on what to include in a model inventory, as well as determining risk appetite, risk tiering, roles and responsibilities, and model life-cycle controls, not to mention the associated model-validation practices. The good news is that many banks will not need entirely new model-validation frameworks. Existing ones can be fitted for purpose with some well-targeted enhancements.
New Risk mitigation exercises for ML models
There is no shortage of news headlines revealing the unintended consequences of new machine-learning models. Algorithms that created a negative feedback loop were blamed for the “flash crash” of the British pound by 6 percent in 2016, for example, and it was reported that a self-driving car tragically failed to properly identify a pedestrian walking her bicycle across the street. The cause of the risks that materialized in these machine-learning models is the same as the cause of the amplified risks that exist in all machine-learning models, whatever the industry and application: increased model complexity. Machine-learning models typically act on vastly larger data sets, including unstructured data such as natural language, images, and speech. The algorithms are typically far more complex than their statistical counterparts and often require design decisions to be made before the training process begins. And machine-learning models are built using new software packages and computing infrastructure that require more specialized skills. The response to such complexity does not have to be overly complex, however. If properly understood, the risks associated with machine-learning models can be managed within banks’ existing model-validation frameworks
Here are the strategic approaches for enterprises to ensure that that the specific risks associated with machine learning are addressed :
Demystification of “Black Boxes” : Machine-learning models have a reputation of being “black boxes.” Depending on the model’s architecture, the results it generates can be hard to understand or explain. One bank worked for months on a machine-learning product-recommendation engine designed to help relationship managers cross-sell. But because the managers could not explain the rationale behind the model’s recommendations, they disregarded them. They did not trust the model, which in this situation meant wasted effort and perhaps wasted opportunity. In other situations, acting upon (rather than ignoring) a model’s less-than-transparent recommendations could have serious adverse consequences.
The degree of demystification required is a policy decision for banks to make based on their risk appetite. They may choose to hold all machine-learning models to the same high standard of interpretability or to differentiate according to the model’s risk. In USA, models that determine whether to grant credit to applicants are covered by fair-lending laws. The models therefore must be able to produce clear reason codes for a refusal. On the other hand, banks might well decide that a machine-learning model’s recommendations to place a product advertisement on the mobile app of a given customer poses so little risk to the bank that understanding the model’s reasons for doing so is not important. Validators need also to ensure that models comply with the chosen policy. Fortunately, despite the black-box reputation of machine-learning models, significant progress has been made in recent years to help ensure their results are interpretable. A range of approaches can be used, based on the model class:
Linear and monotonic models (for example, linear-regression models): linear coefficients help reveal the dependence of a result on the output. Nonlinear and monotonic models, (for example, gradient-boosting models with monotonic constraint): restricting inputs so they have either a rising or falling relationship globally with the dependent variable simplifies the attribution of inputs to a prediction. Nonlinear and nonmonotonic (for example, unconstrained deep-learning models): methodologies such as local interpretable model-agnostic explanations or Shapley values help ensure local interpretability.
Bias : A model can be influenced by four main types of bias: sample, measurement, and algorithm bias, and bias against groups or classes of people. The latter two types, algorithmic bias and bias against people, can be amplified in machine-learning models. For example, the random-forest algorithm tends to favor inputs with more distinct values, a bias that elevates the risk of poor decisions. One bank developed a random-forest model to assess potential money-laundering activity and found that the model favored fields with a large number of categorical values, such as occupation, when fields with fewer categories, such as country, were better able to predict the risk of money laundering.
To address algorithmic bias, model-validation processes should be updated to ensure appropriate algorithms are selected in any given context. In some cases, such as random-forest feature selection, there are technical solutions. Another approach is to develop “challenger” models, using alternative algorithms to benchmark performance. To address bias against groups or classes of people, banks must first decide what constitutes fairness. Four definitions are commonly used, though which to choose may depend on the model’s use: Demographic blindness: decisions are made using a limited set of features that are highly uncorrelated with protected classes, that is, groups of people protected by laws or policies. Demographic parity: outcomes are proportionally equal for all protected classes. Equal opportunity: true-positive rates are equal for each protected class. Equal odds: true-positive and false-positive rates are equal for each protected class. Validators then need to ascertain whether developers have taken the necessary steps to ensure fairness. Models can be tested for fairness and, if necessary, corrected at each stage of the model-development process, from the design phase through to performance monitoring.
Feature engineering : is often much more complex in the development of machine-learning models than in traditional models. There are three reasons why. First, machine-learning models can incorporate a significantly larger number of inputs. Second, unstructured data sources such as natural language require feature engineering as a preprocessing step before the training process can begin. Third, increasing numbers of commercial machine-learning packages now offer so-called AutoML, which generates large numbers of complex features to test many transformations of the data. Models produced using these features run the risk of being unnecessarily complex, contributing to overfitting. For example, one institution built a model using an AutoML platform and found that specific sequences of letters in a product application were predictive of fraud. This was a completely spurious result caused by the algorithm’s maximizing the model’s out-of-sample performance.
In feature engineering, banks have to make a policy decision to mitigate risk. They have to determine the level of support required to establish the conceptual soundness of each feature. The policy may vary according to the model’s application. For example, a highly regulated credit-decision model might require that every individual feature in the model be assessed. For lower-risk models, banks might choose to review the feature-engineering process only: for example, the processes for data transformation and feature exclusion. Validators should then ensure that features and/or the feature-engineering process are consistent with the chosen policy. If each feature is to be tested, three considerations are generally needed: the mathematical transformation of model inputs, the decision criteria for feature selection, and the business rationale. For instance, a bank might decide that there is a good business case for using debt-to-income ratios as a feature in a credit model but not frequency of ATM usage, as this might penalize customers for using an advertised service.
Hyper parameters : Many of the parameters of machine-learning models, such as the depth of trees in a random-forest model or the number of layers in a deep neural network, must be defined before the training process can begin. In other words, their values are not derived from the available data. Rules of thumb, parameters used to solve other problems, or even trial and error are common substitutes. Decisions regarding these kinds of parameters, known as hyper parameters, are often more complex than analogous decisions in statistical modeling. Not surprisingly, a model’s performance and its stability can be sensitive to the hyper parameters selected. For example, banks are increasingly using binary classifiers such as support-vector machines in combination with natural-language processing to help identify potential conduct issues in complaints. The performance of these models and the ability to generalize can be very sensitive to the selected kernel function.Validators should ensure that hyper parameters are chosen as soundly as possible. For some quantitative inputs, as opposed to qualitative inputs, a search algorithm can be used to map the parameter space and identify optimal ranges. In other cases, the best approach to selecting hyperparameters is to combine expert judgment and, where possible, the latest industry practices.
Production readiness : Traditional models are often coded as rules in production systems. Machine-learning models, however, are algorithmic, and therefore require more computation. This requirement is commonly overlooked in the model-development process. Developers build complex predictive models only to discover that the bank’s production systems cannot support them. One US bank spent considerable resources building a deep learning–based model to predict transaction fraud, only to discover it did not meet required latency standards. Validators already assess a range of model risks associated with implementation. However, for machine learning, they will need to expand the scope of this assessment. They will need to estimate the volume of data that will flow through the model, assessing the production-system architecture (for example, graphics-processing units for deep learning), and the runtime required.
Dynamic model calibration : Some classes of machine-learning models modify their parameters dynamically to reflect emerging patterns in the data. This replaces the traditional approach of periodic manual review and model refresh. Examples include reinforcement-learning algorithms or Bayesian methods. The risk is that without sufficient controls, an overemphasis on short-term patterns in the data could harm the model’s performance over time. Banks therefore need to decide when to allow dynamic recalibration. They might conclude that with the right controls in place, it is suitable for some applications, such as algorithmic trading. For others, such as credit decisions, they might require clear proof that dynamic recalibration outperforms static models. With the policy set, validators can evaluate whether dynamic recalibration is appropriate given the intended use of the model, develop a monitoring plan, and ensure that appropriate controls are in place to identify and mitigate risks that might emerge. These might include thresholds that catch material shifts in a model’s health, such as out-of-sample performance measures, and guardrails such as exposure limits or other, predefined values that trigger a manual review.
Banks will need to proceed gradually. The first step is to make sure model inventories include all machine learning–based models in use. One bank’s model risk-management function was certain the organization was not yet using machine-learning models, until it discovered that its recently established innovation function had been busy developing machine-learning models for fraud and cyber security.
From here, validation policies and practices can be modified to address machine-learning-model risks, though initially for a restricted number of model classes. This helps build experience while testing and refining the new policies and practices. Considerable time will be needed to monitor a model’s performance and finely tune the new practices. But over time banks will be able to apply them to the full range of approved machine-learning models, helping companies mitigate risk and gain the confidence to start harnessing the full power of machine learning.
(AIQRATE, A bespoke global AI advisory and consulting firm. A first in its genre, AIQRATE provides strategic AI advisory services and consulting offerings across multiple business segments to enable clients on their AI powered transformation & innovation journey and accentuate their decision making and business performance.
AIQRATE works closely with Boards, CXOs and Senior leaders advising them on navigating their Analytics to AI journey with the art of possible or making them jump start to AI progression with AI@scale approach followed by consulting them on embedding AI as core to business strategy within business functions and augmenting the decision-making process with AI. We have proven bespoke AI advisory services to enable CXO’s and Senior Leaders to curate & design building blocks of AI strategy, embed AI@scale interventions and create AI powered organizations. AIQRATE’s path breaking 50+ AI consulting frameworks, assessments, primers, toolkits and playbooks enable Indian & global enterprises, GCCs, Startups, VC/PE firms, and Academic Institutions enhance business performance and accelerate decision making.
Visit www.aiqrate.ai to experience our AI advisory services & consulting offerings
Webinar on AI Strategy at BITS Pilani
Continuing with the webinar series for the batch of 2021 & 2022 along with our noteworthy alumni, BITS Pilani called-in AIQRATE for a session is on ‘AI Strategy : The new next in Transformation and Innovation’ by Mr Sameer Dhanrajani , on 26th August 2020.
Personal Data Sharing & Protection: Strategic relevance from India’s context
India’s Investments in the digital financial infrastructure—known as “India Stack”—have sped up the large-scale digitization of people’s financial lives. As more and more people begin to conduct transactions online, questions have emerged about how to provide millions of customers adequate data protection and privacy while allowing their data to flow throughout the financial system. Data-sharing among financial services providers (FSPs) can enable providers to more efficiently offer a wider range of financial products better tailored to the needs of customers, including low-income customers.
There are several operational and coordination challenges across these three types of entities: FIPs, FIUs, and AAs. There are also questions around the data-sharing business model of AAs. Since AAs are additional players, they generate costs that must be offset by efficiency gains in the system to mitigate overall cost increases to customers. It remains an open question whether AAs will advance financial inclusion, how they will navigate issues around digital literacy and smartphone access, how the limits of a consent-based model of data protection and privacy play out, what capacity issues will be encountered among regulators and providers, and whether a competitive market of AAs will emerge given that regulations and interoperability arrangements largely define the business.
Account Aggregators (AA’s):
ACCOUNT AGGREGATORS (AAs) is one of the new categories of non banking financial companies (NBFCs) to figure into India Stack—India’s interconnected set of public and nonprofit infrastructure that supports financial services. India Stack has scaled considerably since its creation in 2009, marked by rapid digitization and parallel growth in mobile networks, reliable data connectivity, falling data costs, and continuously increasing smartphone use. Consequently, the creation, storage, use, and analyses of personal data have become increasingly relevant. Following an “open banking “approach, the Reserve Bank of India (RBI) licensed seven AAs in 2018 to address emerging questions around how data can be most effectively leveraged to benefit individuals while ensuring appropriate data protection and privacy, with consent being a key element in this. RBI created AAs to address the challenges posed by the proliferation of data by enabling data-sharing among financial institutions with customer consent. The intent is to provide a method through which customers can consent (or not) to a financial services provider accessing their personal data held by other entities. Providers are interested in these data, in part, because information shared by customers, such as bank statements, will allow providers to better understand customer risk profiles. The hypothesis is that consent-based data-sharing will help poorer customers qualify for a wider range of financial products—and receive financial products better tailored to their needs.
Data Sharing Model : The new perspective:
Paper based data collection is inconvenient , time consuming and costly for customers and providers. Where models for digital-sharing exist, they typically involve transferring data through intermediaries that are not always secure or through specialized agencies that offer little protection for customers. India’s consent-based data-sharing model provides a digital framework that enables individuals to give and withdraw consent on how and how much of their personal data are shared via secure and standardized channels. India’s guiding principles for sharing data with user consent—not only in the financial sector— are outlined in the National Data Sharing and Accessibility Policy (2012) and the Policy for Open Application Programming Interfaces for the Government of India. The Information Technology Act (2000) requires any entity that shares sensitive personal data to obtain consent from the user before the information is shared. The forthcoming Personal Data Protection Bill makes it illegal for institutions to share personal data without consent.
- Identifier : Specifies entities involved in the transaction: who is requesting the data, who is granting permission, who is providing the data, and who is recording consent.
- Data : Describes the type of data being accessed and the permissions for use of the data. Three types of permissions are available: view (read only), store, and query (request for specific data). The artifact structure also specifies the data that are being shared, date range for which they are being requested, duration of storage by the consumer, and frequency of access.
- Purpose : Describes end use, for example, to compute a loan offer.
- Log : Contains logs of who asked for consent, whether it was granted or not, and data flows.
- Digital signature : Identifies the digital signature and digital ID user certificate used by the provider to verify the digital signature. This allows providers to share information in encrypted form
The Approach :
THE AA consent based data sharing model mediates the flow of data between producers and users of data, ensuring that sharing data is subject to granular customer consent. AAs manage only the consent and data flow for the benefit of the consumer, mitigating the risk of an FIU pressuring consumers to consent to access to their data in exchange for a product or service. However, AAs, as entities that sit in the middle of this ecosystem, come with additional costs that will affect the viability of the business model and the cost of servicing consumers. FIUs most likely will urge consumers to go directly to an AA to receive fast, efficient, and low-cost services. However, AAs ultimately must market their services directly to the consumer. While AA services are not an easy sell, the rising levels of awareness among Indian consumers that their data are being sold without their consent or knowledge may give rise to the initial wave of adopters. While the AA model is promising, it remains to be seen how and when it will have a direct impact on the financial lives of consumers.
Differences between Personal Data Protection & GDPR ?
There are some major differences between the two.
First, the bill gives India’s central government the power to exempt any government agency from the bill’s requirements. This exemption can be given on grounds related to national security, national sovereignty, and public order.
While the GDPR offers EU member states similar escape clauses, they are tightly regulated by other EU directives. Without these safeguards, India’s bill potentially gives India’s central government the power to access individual data over and above existing Indian laws such as the Information Technology Act of 2000, which dealt with cyber crime and e-commerce.
Second, unlike the GDPR, India’s bill allows the government to order firms to share any of the non personal data they collect with the government. The bill says this is to improve the delivery of government services. But it does not explain how this data will be used, whether it will be shared with other private businesses, or whether any compensation will be paid for the use of this data.
Third, the GDPR does not require businesses to keep EU data within the EU. They can transfer it overseas, so long as they meet conditions such as standard contractual clauses on data protection, codes of conduct, or certification systems that are approved before the transfer.
The Indian bill allows the transfer of some personal data, but sensitive personal data can only be transferred outside India if it meets requirements that are similar to those of the GDPR. What’s more, this data can only be sent outside India to be processed; it cannot be stored outside India. This will create technical issues in delineating between categories of data that have to meet this requirement, and add to businesses’ compliance costs.
AI Strategy: The Epiphany of Digital Transformation
In the past months due to lockdowns and WFH, enterprises have got an epiphany of massive shifts of business and strategic models for staying relevant and solvent. Digital transformation touted as the biggest strategic differentiation and competitive advantages for enterprises faced a quintessential inertia of mass adoption in the legacy based enterprises and remained more on business planning slides than in full implementation. However, Digital Transformation is not about aggregation of exponential technologies and adhoc use cases or stitching alliances with deep tech startups. The underpinning of Digital transformation is AI and how AI strategy has become the foundational aspect of accomplishing digital transformation for enterprises and generating tangible business metrics. But before we get to the significance of AI strategy in digital transformation, we need to understand the core of digital transformation itself. Because digital transformation will look different for every enterprise, it can be hard to pinpoint a definition that applies to all. However, in general terms: we define digital transformation as the integration of core areas of business resulting in fundamental changes to how businesses operate and how they deliver value to customers.
Though, in specific terms digital transformation can take a very interesting shape according to the business moment in question. From a customer’s point of view, “Digital transformation closes the gap between what digital customers already expect and what analog businesses actually deliver.”
Does Digital Transformation really mean bunching exponential technologies? I believe that digital transformation is first and foremost a business transformation. Digital mindset is not only about new age technology, but about curiosity, creativity, problem-solving, empathy, flexibility, decision-making and judgment, among others. Enterprises needs to foster this digital mindset, both within its own boundaries and across the company units. The World Economic Forum lists the top 10 skills needed for the fourth industrial revolution. None of them is totally technical. They are, rather, a combination of important soft skills relevant for the digital revolution. You don’t need to be a technical expert to understand how technology will impact your work. You need to know the foundational aspects, remain open-minded and work with technology mavens. Digital Transformation is more about cultural change that requires enterprises to continually challenge the status quo, experiment often, and get comfortable with failure. The most likely reason for business to undergo digital transformation is the survival & relevance issue. Businesses mostly don’t transform by choice because it is expensive and risky. Businesses go through transformation when they have failed to evolve. Hence its implementation calls for tough decisions like walking away from long-standing business processes that companies were built upon in favor of relatively new practices that are still being defined.
Business Implementation aspects of Digital Transformation
Disruption in digital business implies a more positive and evolving atmosphere, instead of the usual negative undertones that are attached to the word. According to the MIT Center for Digital Business, “Companies that have embraced digital transformation are 26 percent more profitable than their average industry competitors and enjoy a 12 percent higher market valuation.” A lot of startups and enterprises are adopting an evolutionary approach in transforming their business models itself, as part of the digital transformation. According to Mckinsey, One-third of the top 20 firms in industry segments will be disrupted by new competitors within five years.
The various Business Models being adopted in Digital Transformation era are:
- The Subscription Model (Netflix, Dollar Shave Club, Apple Music) Disrupts through “lock-in” by taking a product or service that is traditionally purchased on an ad hoc basis, and locking-in repeat custom by charging a subscription fee for continued access to the product/service
- The Freemium Model (Spotify, LinkedIn, Dropbox) Disrupts through digital sampling, where users pay for a basic service or product with their data or ‘eyeballs’, rather than money, and then charging to upgrade to the full offer. Works where marginal cost for extra units and distribution are lower than advertising revenue or the sale of personal data
- The Free Model (Google, Facebook) Disrupts with an ‘if-you’re-not-paying-for-the-product-you-are-the-product’ model that involves selling personal data or ‘advertising eyeballs’ harvested by offering consumers a ‘free’ product or service that captures their data/attention
- The Marketplace Model (eBay, iTunes, App Store, Uber, Airbnb) Disrupts with the provision of a digital marketplace that brings together buyers and sellers directly, in return for a transaction or placement fee or commission
- The Access-over-Ownership Model (Zipcar, Peer buy) Disrupts by providing temporary access to goods and services traditionally only available through purchase. Includes ‘Sharing Economy’ disruptors, which takes a commission from people monetizing their assets (home, car, capital) by lending them to ‘borrowers’
- The Hypermarket Model (Amazon, Apple) Disrupts by ‘brand bombing’ using sheer market power and scale to crush competition, often by selling below cost price
- The Experience Model (Tesla, Apple) Disrupts by providing a superior experience, for which people are prepared to pay
- The Pyramid Model (Amazon, Microsoft, Dropbox) Disrupts by recruiting an army of resellers and affiliates who are often paid on a commission-only mode
- The On-Demand Model (Uber, Operator, TaskRabbit) Disrupts by monetizing time and selling instant-access at a premium. Includes taking a commission from people with money but no time who pay for goods and services delivered or fulfilled by people with time but no money
- The Ecosystem Model (Apple, Google) Disrupts by selling an interlocking and interdependent suite of products and services that increase in value as more are purchased. Creates consumer dependency
Since Digital Transformation and its manifestation into various business models are being fast adopted by startups, there are providing tough competition to incumbent corporate houses and large enterprises. Though enterprises are also looking forward to digitally transform their enterprise business, the scale and complexity makes it difficult and resource consuming activity. It has imperatively invoked the enterprises to bring certain strategy to counter the cannibalizing effect in the following ways:
- The Block Strategy. Using all means available to inhibit the disruptor. These means can include claiming patent or copyright infringement, erecting regulatory hurdles, and using other legal barriers.
- The Milk Strategy. Extracting the most value possible from vulnerable businesses while preparing for the inevitable disruption
- The Invest in Disruption Model. Actively investing in the disruptive threat, including disruptive technologies, human capabilities, digitized processes, or perhaps acquiring companies with these attributes
- The Disrupt the Current Business Strategy. Launching a new product or service that competes directly with the disruptor, and leveraging inherent strengths such as size, market knowledge, brand, access to capital, and relationships to build the new business
- The Retreat into a Strategic Niche Strategy. Focusing on a profitable niche segment of the core market where disruption is less likely to occur (e.g. travel agents focusing on corporate travel, and complex itineraries, book sellers and publishers focusing on academia niche)
- The Redefine the Core Strategy. Building an entirely new business model, often in an adjacent industry where it is possible to leverage existing knowledge and capabilities (e.g. IBM to consulting, Fujifilm to cosmetics)
- The Exit Strategy. Exiting the business entirely and returning capital to investors, ideally through a sale of the business while value still exists (e.g. MySpace selling itself to Newscorp)
The curious evolution of AI and its relevance in digital transformation
So here’s an interesting question, AI has been around for more than 60 years, then why is it that it is only gaining traction with the advent of digital? The first practical application of such “machine intelligence” was introduced by Alan Turing, British mathematician and WWII code-breaker, in 1950. He even created the Turing test, which is still used today, as a benchmark to determine a machine’s ability to “think” like a human.The biggest differences between AI then and now are Hardware limitations, access to data, and rise of machine learning.
Hardware limitations led to the non-sustenance of AI adoption till late 1990s. There were many instances where the scope and opportunity of AI led transformation was identified and appreciated by implementation saw more difficult circumstances. The field of AI research was founded at a workshop held on the campus of Dartmouth College during the summer of 1956. But Eventually it became obvious that they had grossly underestimated the difficulty of the project due to computer hardware limitations. The U.S. and British Governments stopped funding undirected research into artificial intelligence, leading to years known as an “AI winter”.
In another example, again in 1980, a visionary initiative by the Japanese Government inspired governments and industry to provide AI with billions of dollars, but by the late 80s the investors became disillusioned by the absence of the needed computer power (hardware) and withdrew funding again. Investment and interest in AI boomed in the first decades of the 21st century, when machine learning was successfully applied to many problems in academia and industry due to the presence of powerful computer hardware. Teaming this with the rise in digital, leading to an explosion of data and adoption of data generation in every aspect of business, made it highly convenient for AI to not only be adopted but to evolve to more accurate execution.
The Core of Digital Transformation: AI Strategy
According to McKinsey, by 2023, 85 percent of all digital transformation initiatives will be embedded with AI strategy at its core. Due to radical computational power, near-endless amounts of data, and unprecedented advances in ML algorithms, AI strategy will emerge as the most disruptive business scenario, and its manifestation into various trends that we see and will continue to see, shall drive the digital transformation as we understand it. The following will the future forward scenarios of AI strategy becoming core to digital transformation:
AI’s growing entrenchment: This time, the scale and scope of the surge in attention to AI is much larger than before. For starters, the infrastructure speed, availability, and sheer scale has enabled bolder algorithms to tackle more ambitious problems. Not only is the hardware faster, sometimes augmented by specialized arrays of processors (e.g., GPUs), it is also available in the shape of cloud services , data farms and centers
Geography, societal Impact: AI adoption is reaching institutions outside of the industry. Lawyers will start to grapple with how laws should deal with autonomous vehicles; economists will study AI-driven technological unemployment; sociologists will study the impact of AI-human relationships. This is the world of the future and the new next.
Artificial intelligence will be democratized: As per the results of a recent Forrester study , it was revealed that 58 percent of professionals researching artificial intelligence ,only 12 percent are actually using an AI system. Since AI requires specialized skills or infrastructure to implement, Companies like Facebook have realized this and are already doing all they can to simplify the implementation of AI and make it more accessible. Cloud platforms like Google APIs, Microsoft Azure, AWS are allowing developers to create intelligent apps without having to set up or maintain any other infrastructure.
Niche AI will Grow: By all accounts, 2020 & beyond won’t be for large, general-purpose AI systems. Instead, there will be an explosion of specific, highly niche artificial intelligence adoption cases. These include autonomous vehicles (cars and drones), robotics, bots (consumer-orientated such as Amazon Echo , and industry specific AI (think finance, health, security etc.).
Continued Discourse on AI ethics, security & privacy: Most AI systems are immensely complex sponges that absorb data and process it at tremendous rates. The risks related to AI ethics, security and privacy are real and need to be addressed through consideration and consensus. Sure, it’s unlikely that these problems will be solved in 2020, but as long as the conversation around these topics continues, we can expect at least some headway.
Algorithm Economy: With massive data generation using flywheels, there will be an economy created for algorithms, like a marketplace for algorithms. The engineers, data scientists, organizations, etc. will be sharing algorithms for using the data to extract required information set.
Where is AI Heading in the Digital Road?While much of this is still rudimentary at the moment, we can expect sophisticated AI to significantly impact our everyday lives. Here are four ways AI might affect us in the future:
Humanizing AI: AI will grow beyond a “tool” to fill the role of “co-worker.” Most AI software is too hidden technologically to significantly change the daily experience for the average worker. They exist only in a back end with little interface with humans. But several AI companies combine advanced AI with automation and intelligent interfaces that drastically alter the day to day workflow for workers
Design Thinking & behavioral science in AI: We will witness Divergence from More Powerful Intelligence To More Creative Intelligence. There have already been attempts to make AI engage in creative efforts, such as artwork and music composition. we’ll see more and more artificial intelligence designing artificial intelligence, resulting in many mistakes, plenty of dead ends, and some astonishing successes.
Rise of Cyborgs: As augmented AI is already the mainstream thinking; the future might hold witness to perfect culmination of man-machine augmentation. AI augmented to humans, intelligently handling operations which human cannot do, using neural commands.
AI Oracle : AI might become so connected with every aspect of our lives, processing though every quanta of data from every perspective that it would perfectly know how to raise the overall standard of living for the human race. People would religiously follow its instructions (like we already follow GPS navigations) leading to leading to an equation of dependence closer to devotion.
The Final Word
Digital business transformation is the ultimate challenge in change management. It impacts not only industry structures and strategic positioning, but it affects all levels of an organization (every task, activity, process) and even its extended supply chain. Hence to brace Digital led disruption, one has to embrace AI-led strategy. Organizations that deploy AI strategically will ultimately enjoy advantages ranging from cost reductions and higher productivity to top-line benefits such as increasing revenue and profits, richer customer experiences, and working-capital optimization.
( AIQRATE, A bespoke global AI advisory and consulting firm. A first in its genre, AIQRATE provides strategic AI advisory services and consulting offerings across multiple business segments to enable clients navigate their AI powered transformation, innovation & revival journey and accentuate their decision making and business performance.
AIQRATE works closely with Boards, CXOs and Senior leaders advising them on their Analytics to AI journey construct with the art of possible AI roadmap blended with a jump start approach to AI driven transformation with AI@scale centric strategy; AIQRATE also consults on embedding AI as core to business strategy within business processes & functions and augmenting the overall decision-making capabilities. Our bespoke AI advisory services focus on curating & designing building blocks of AI strategy, embed AI@scale interventions and create AI powered organizations.
AIQRATE’s path breaking 50+ AI consulting frameworks, methodologies, primers, toolkits and playbooks crafted by seasoned and proven AI strategy advisors enable Indian & global enterprises, GCCs, Startups, SMBs, VC/PE firms, and Academic Institutions enhance business performance & ROI and accelerate decision making capability. AIQRATE also provide advisory support to Technology companies, business consulting firms, GCCs, AI pure play outfits on curating discerning AI capabilities, solutions along with differentiated GTM and market development strategies.
Experiential Masterclass: AI Strategy for Enterprise Decision Making
The most awaited experiential masterclass on AI Strategy for Enterprise Decision Making was held on Saturday, June 13th 2020 with global participants.
This bespoke and experiential masterclass will be delivered by a seasoned AI evangelist and business builder. Sameer has a proven capability of scaling AI business practices & building AI CoE’s, has consulted with several global & Indian enterprises, GCC’s on AI strategy & transformation, executed 3000+ AI & Analytics consulting engagements. The Masterclass will compel the participants to cogitate towards developing AI strategies in conjunction with looking at developing frameworks and action plans for leveraging AI capabilities within their organizations and business functions for inculcating Transformation, Innovation and Disruption dynamics within their organizations. The participants will also be showcased with topical scenarios, best practices and global trends in AI arena.
The need to have an AI strategy in crisis : Reset & Revive
With the global lock down caused by the COVID-19 and the unforeseen loss of business momentum , the luxury of time now seems to have disappeared completely. Businesses that once mapped strategy planning in one- three-year phases must now reset and scale their strategic initiatives in a matter of days or weeks. In one of the survey initiated by Harvard university , about 70 percent of top fortune 1000 companies senior executives said the pandemic is likely to accelerate the pace of their business transformation. The acceleration is evident already across sectors and geographies. Consider how multiple banks have swiftly migrated physical channels online. How healthcare providers have moved rapidly into tele-health, insurers into self-service claims assessment, and retailers into contactless shopping and delivery.
The COVID-19 crisis seemingly provides a sudden glimpse into a future world, one in which artificial intelligence has become central to every interaction, forcing both enterprises and individuals further up the adoption curve almost overnight. A world in which digital channels become the primary customer-engagement model, and automated processes become a primary driver of productivity—and the basis of flexible, transparent, and stable supply chains. A world in which agile ways of working are a prerequisite to meeting seemingly daily changes to customer behavior. This being powered by a robust AI driven algorithmic engines . If a silver lining can be found, it might be in the falling barriers to improvisation and experimentation that have emerged among customers, markets, regulators, and organizations. In this unique moment, enterprises can learn and progress more quickly than ever before. The ways they reset and revive post crisis will deeply influence their performance in tomorrow’s transformative world, providing the opportunity to retain greater agility as well as closer ties with customers, employees, and suppliers. Those that are successfully able to make gains will likely be more successful during recovery and beyond.
Now is the time to reassess business strategy and curate AI strategy core to the business models & processes—to provide near-term readiness to employees, customers, and the broad set of stakeholders to which businesses are increasingly responsible and those that position you for a post crisis world. In this world, some things will snap back to previous form, while others will be forever changed. Playing it safe now, understandable as it might feel to do so, is often the worst option.
A Black Swan event demands new strategic approaches : AI Strategy comes to the rescue
Every enterprise knows the virtues of how AI pilots new business models in “normal” times, but very have implemented AI strategy @scale and velocity suddenly required by the COVID-19 crisis. That’s because in normal times, the customer and market penalties for widespread “test and learn” can seem too high, and the enterprises obstacles too steep. Shareholders of public companies demand immediate returns. Finance departments keep tight hold of the funds needed to move new initiatives forward quickly. Customers are often slow to adjust to new ways of doing things, with traditional adoption curves reflecting this inherent inertia. And organizational culture, with its own siloes, hinders agility and collaboration. As a result, enterprises often experiment at a pace that fails to match the rate of change around them, slowing their ability to learn fast enough to keep up. Additionally, they rarely embrace the acceleration needed to move quickly from piloting initiatives to scaling the successful ones, even though analyst studies have shown that swift moves to curate AI strategy early and at scale, combined with a sizeable allocation of resources against AI implementation , correlate highly with value creation As the COVID-19 crisis forces your customers, employees, and supply chains into digital channels and new ways of working, now is the time to ask : Does my enterprise have an AI strategy to reimagine customer experiences , innovate new products & services and transform my business for competitive advantage ? Strange as it may seem, right now, in a moment of crisis, is precisely the time to boldly advance your move to curate an AI strategy .
AI Strategy Curation : Strategic Focus Areas :
Crafting an AI strategy goes beyond building light weight , beta mode algorithms , pursuing adhoc business problems for driving AI engagements or cobbling up together a bunch of AI geeks ; it requires a strategic approach driven by boards , CXOs’ , business leaders and decision makers to focus on the following key areas :
1.Craft Novel Business capabilities embedded with AI
By now you have built your contingency response model and insights hub; you need to coordinate your crisis response. This insights hub provides a natural gathering point for crucial strategic information, helping you stay close to the quickly evolving needs of core customer segments, and the ways in which competitors and markets are moving to meet them. Mapping these changes helps address immediate risks, to be sure, but it also affords looking forward in time at bigger issues and opportunities—those that could drive significant disruption as the crisis continues. Just as AI has disrupted business models and value chains in the past, the COVID-19 crisis will set similar “ecosystem”-level changes in motion—not just changes in economics but new ways of serving customers and working with suppliers across in a new ecosystem. In the immediate term, for example, most enterprises are looking for virtual capabilities for their previously physical offerings, or at least new ways of making them accessible with minimal physical contact. The new offerings that result can often involve new partnerships or the need to access new platforms and digital marketplaces in which your company has yet to participate. As you engage with new partners and platforms, look for opportunities to move beyond your organization’s comfort zones, while getting visibility into the places you can confidently invest valuable time, people, and funds to their best effect. AI based strategy that involves building recommended intelligence systems, reasoning and intuition to address complex problems and explore ideal future states, will be crucial.
2. Embed AI into your core business model
Going beyond comfort zones requires taking an end-to-end view of your business and operating models. Even though your resources are necessarily limited, the experience of leading enterprises suggests that focusing on embedding AI in to the areas that touch more of the core of your business will give you the best chance of success, in both the near and the longer term, than will making minor improvements to noncore areas. Enterprises that make minor changes to the edges of their business model nearly always falter in their business goals. Tinkering leads to returns on investment below the cost of capital and to changes that are too small to match the external pace of disruption. Enterprises that rapidly adopts embedding AI driven algorithms and using those to redefine their business at scale have been outperforming their peers. This will be increasingly true as companies deal with large amounts of data in a rapidly evolving landscape and look to make rapid, accurate course corrections compared with their peers. On a short term basis , this may mean , opening up business models for introspection , however, embedding AI into the core business areas : marketing , sales , supply chain , finance will radically change your enterprise’s ability to derive insights & intelligence.
3. Reset your business strategies with AI
No enterprise can accelerate the delivery of all its strategic imperatives without looking to M&A to speed them along. This is particularly true with AI strategy, where M&A can help companies gain talent and build capabilities, even as it offers access to new products, services, and solutions, and to new market and customer segments. More broadly, we know from research from previous black swan events that enterprises that invest when valuations are low outperform those that do not. In more normal times, one of the main challenges enterprises face in their AI led transformations and adoption is the need to acquire AI talent and capabilities through acquisitions of startups that are typically valued at multiples that capital markets might view as dilutive to the acquirer. The current downturn could remove this critical roadblock, especially with enterprises temporarily free from the tyranny of quarterly earnings expectations.
In the next part of the series , I will elaborate on the steps and interventions that are required to craft & curate an AI strategy . Stay Tuned…..
Strategic perspectives for India to attain AI supremacy
The strategic perspectives provided herein will provide you crucial overview of the AI’s increasing prevalence amongst Indian industry, government and peripheral ecosystem and the significant impact AI will generate for India in the coming years and the possible strategic considerations that India needs to initiate to attain AI supremacy. The ensuing details also highlights the relative comparison amongst India, China and USA on the steady progress being done in AI adoption. VC firms, PE funds and investors attempting to understand where to target investment, what offerings and capabilities would lead to better performance and gains, and how to capitalize on AI opportunities, it’s crucial for them to understand the International economic potential of AI for now and projections in the coming years. Cutting across all these strategic considerations is how to build responsible AI operating models and keep it transparent enough to maintain the confidence of customers and wider stakeholders.
International AI Capitalization Report – China & NA Leads, India hot in the heels
Without doubt, AI is going to be a big game changer in the international setting. A previous set of reports from multiple analysts concluded that AI could contribute up to $15.7 trillion to the global economy in 2030, more than the current output of China and India combined. Of this, $6.6 trillion is likely to come from increased productivity and $9.1 trillion is likely to come from consumption-side effects. Global GDP will be up to 14% higher in 2030 as a result of the accelerating development and take-up of AI from the standpoint of direct economic impact of AI, China and USA will have greatest gains in GDP. Even though USA will reach its peak of AI led growth faster due to huge opportunities in parallel technologies implementations and advanced customer readiness for AI.
China, on the other hand will have a slower but stable rise in GDP gains, post COVID 19 because a large portion of Chinese GDP comes from manufacturing, a sector which is highly susceptible to AI disruption in its operation, and also a higher rate of capital re-investment within Chinese economy compared to EU and USA. As productivity in China eventually catches up with USA , USA will focus more on importing AI-enabled products from China due to economically cheap alternative China provides. Hence by 2030, China will see much larger impact in its GDP.
Is the Differential for Developing countries like India too steep in catching up with AI? – AI is still at its early stages, which means that irrespective of the fact that the exponential technology landscape is skewed towards the developed economies as compared to developing, the developing economies and their markets could still lead the developed markets from AI standpoint. This makes countries like India, with a strong focus in Technology sector, a strong contender.
The economic impact of AI in GDP for India ,will be driven by:
- Productivity gains from businesses automating processes (including use of robots and autonomous vehicles).
- Productivity gains from businesses augmenting their existing labor force with AI technologies (assisted and augmented intelligence).
- Increased consumer demand resulting from the availability of personalized and/or higher-quality AI-enhanced products and services.
The consumer revolution set off by AI opens the way for massive disruption as both established businesses and new entrants drive innovation and develop new business models. A key part of the impact of AI will come from its ability to make the most of parallel developments such as 5G connectivity.
India’s Macroeconomic Landscape of AI
India is already way ahead of many other countries in implementing artificial intelligence (AI). More than 40% of the enterprises are going beyond pilot and test projects and adopting the technology at a larger scale coupled with 1400+ global capability centers that have become frontiers in pushing AI led innovation and transformation for their parent organizations. The Indian government’s Digital India initiative, too, has created a favorable regulatory environment for increased use of AI.
Recipe for AI Success in India – Digital Deluge & Data Detonation
As India undergoes rapid digital transformation, data centers and the intelligence behind the data collected will enable the government and industry to make effective decisions based on algorithms. This means increasing opportunities for adoption (and investing over) AI in the country.
Intel is betting on Artificial Intelligence (AI) to drive demand for its electronic chips, for which it is aiming to train 15,000 scientists, developers, engineers and students on AI in India over the next one year. The company will host 60 courses under its ‘AI Developer Education Program’. These will train people on ways they can adopt AI for better research, testing or even building of products. Intel is looking at India due to the country’s large base of technical talent. The country is the third largest global site for AI companies. As India’s largest e-commerce marketplace Flip kart is looking to put in use its mammoth pile of data to predict sales of products months in advance. The company is working on an artificial intelligence (AI) solution that will give it an edge over rivals by helping it make smarter decisions in ordering, distribution and pricing products on its platform. Ultimately, the AI system will allow Flip kart to boost efficiency and reduce the cost of products for customers. While rival Amazon, which has around a 10-year head start over Flip kart, is known to have some of the most advanced sales prediction engines, the Indian company has the advantage of having a bigger data set of the country’s online consumer market.
AI Inroads in the Private Sector
AI has now a significant impact in the day to day lives of the regular mass of the country. Now that the Indian IT sector has reached a certain intermediary peak of digitization, the focus, now , is more on automating the repetitive problems and finding more optimized, efficient or refined methods of performing the same tasks, with less time duration and lesser manpower. The result is the standardization of some very critical app based services like virtual assistants, cab aggregators, shopping recommendations etc. This will eventually lead to AI solutions to real world problems.
The AI Startups Sphere of India- Startups are clearly playing a major role in innovating faster than enterprises, which has led to several partnerships. SAP India has invested in Niki.ai, a bot that improves the ordering experience. Then there’s Ractrack.AI, where a bot improves customer engagement and provides insights; it functions as a virtual communications assistant to convert the customer into a client. Racetrack is helping companies turn leads into meaningful engagements by using AI. Another startup, LUCEP, converts all potential queries into leads with their AI engine. The objective is to generate insights from data and simplify customer interaction with a business and also convert them into leads. Indian startups saw $ 10 billion in risk capital being deployed across 1,540 angel and VC/PE deals between January and December 2019. VC/PE firms predict that AI would be key themes to invest in for next few years.
AI in Public Sector– Ripe for Digital Revamp and AI Adoption
A Blue Ocean for AI Investment due to Digital India Initiatives – Though both corporates and startups are making significant inroads in instituting AI in their service architecture and product offerings, and sometimes as part of their core business strategy itself, the challenges in the public sector in instituting AI can be quickly overcome due to huge Digital Movements instituted by the Indian Govt. like Digital India, Skill India and Make in India. This will create a solid bedrock of Data and Digital Footprint which will act as a foundational infrastructure to base AI implementation on, opening a huge blue ocean in public sector, rich for AI investment.
A New Workaround for Regulatory Challenges in Public Sector AI Implementation – One of the peculiar problems the public sector faces in mainstream implementation of AI is the fact that since AI is a continuously self-learning system, capable of analytical or creative decision making and autonomous implementation of actions, who will then be accountable in taking responsibility for its actions, should they turn out to be not so favorable. This is because of the fact that since AI has a degree of autonomous decision making, it makes it difficult to pre-meditate its consequence. The AI systems are meant to augment and enrich the life of the consumers. In such a situation, deciding liability of AI system’s actions will be difficult. Therefore, a lot of deliberation will be required to carefully come to a precise conclusion surrounding implementing these systems with ethical foundation and propriety.
Although many countries like US and some European countries are in the verge of implementing regulations and laws surrounding concepts like driver less vehicles, India still don’t have the regulations sanctioned. This, but need not be a bad news. India is cut to establish a completely revamped legal infrastructure, thereby completely circumventing the need for continuous regulatory intervention. Also, there is a favorable atmosphere in India as far as AI is concerned which will foster a spike in activities in that avenue.
Indian Governance Initiatives – Huge Scope for Investment of AI – As India emerges as a premier destination for AI, scope for investment opens in the governance aspect, in several ways. Governance schemes have a unique trait of the baggage of large volume and large scale implementation need, which can be tackled with Deep learning. For example, in Swachh Bharat Initiative, specifically construction of toilets in rural India, public servants are tasked with uploading images of these toilet constructions to a central server for assessment. Image recognition can be used to target unfinished toilets. It can also be used to identify whether the same official appears in multiple images or if photos were uploaded from a different location other than the intended place. Other initiatives such as the Make in India, Digital India & Skill India can be augmented with AI to deal with scale. The range of application for AI techniques could range from crop insurance schemes, tax fraud detection, and detecting subsidy leakage and defense and security strategy.
An AI system can improve and enrich the agriculture of India by enhancing the bodies like The Department of Agriculture Cooperation and Farmers Welfare, Ministry of Agriculture runs the Kisan Call Centers across the country etc. It can help assist the call center by linking various available information like soil reports from government agencies and link them to the environmental conditions. It will then provide advice on the optimal crop that can be sown in that land pocket. As the need for large scale implementation and monitoring of governance initiative becomes more pronounced, the need for AI becomes absolute and it will open doors to considerable AI investment in the future of India.
Finally, Looking Ahead – A Collaborative Innovation led ecosystem
AI innovations which fall under assisted, augmented and autonomous intelligence will help users understand and decide which level of intelligence is helpful and required in their context, thereby making AI Acceptance easier for the people. At the same time, this AI continuum can be used to understand economic ramifications, usage complexity and decision-making implications. While academia and the private sector conduct research on various AI problems with diverse implications in mind, the public sector with its various digital initiatives (Digital India, Make in India, etc.) can identify areas where parts of the AI continuum can be utilized to increase reach, effectiveness and efficiency, thereby giving direction to AI Innovative Research. A collaborative innovation environment between academia and the private and public sectors will help provide holistic and proactive advisory delivery to the population, for example through public call centers, linking information from various government sources. At the same time, the rich data generated from these interactions can be used to draw deep conclusions. Collaboration between the three pillars could further help get a comprehensive view of problems and find intelligent and innovative ways to increase the efficiency and effectiveness of services delivered to society. India is at a cusp of taking a upward trajectory on establishing AI supremacy ; a strategic roadmap across public, private , SMB’s , Academic and startup sectors will accelerate the path to AI adoption and unleashing new sources of economic output for the country . The journey to attain AI supremacy has begun ……
Lock in winning AI deals : Strategic recommendations for enterprises & GCCs
Artificial Intelligence is unleashing exciting growth opportunities for the enterprises & GCCs , at the same time , they also present challenges and complexities when sourcing, negotiating and enabling the AI deals . The hype surrounding this rapidly evolving space can make it seem as if AI providers hold the most power at the negotiation table. After all, the market is ripe with narratives from analysts stating that enterprises and GCCs failing to embrace and implement AI swiftly run the risk of losing their competitiveness. With pragmatic approach and acknowledgement of concerns and potential risks, it is possible to negotiate mutually beneficial contracts that are flexible, agile and most importantly, scalable. The following strategic choices will help you lock in winning AI deals :
Understand AI readiness & roadmap and use cases
It can be difficult to predict exactly where and how AI can be used in the future as it is constantly being developed, but creating a readiness roadmap and identifying your reckoner of potential use cases is a must. Enterprise and GCC readiness and roadmap will help guide your sourcing efforts for enterprises and GCCs , so you can find the provider best suited to your needs and able to scale with your business use cases. You must also clearly frame your targeted objectives both in your discussions with vendors as well as in the contract. This includes not only a stated performance objective for the AI , but also a definition of what would constitute failure and the legal consequences thereof.
Understand your service provider’s roadmap and ability to provide AI evolution to steady state
Once you begin discussions with AI vendors & providers, be sure to ask questions about how evolved their capabilities and offerings are and the complexity of data sets that were used to train their system along with the implementation use cases . These discussions can uncover potential business and security risks and help shape the questions the procurement and legal teams should address in the sourcing process. Understanding the service provider’s roadmap will also help you decide whether they will be able to grow and scale with you. Gaining insight into the service provider’s growth plans can uncover how they will benefit from your business and where they stand against their competitors. The cutthroat competition among AI rivals means that early adopter enterprises and GCCs that want to pilot or deploy AI@scale will see more capabilities available at ever-lower prices over time. Always mote that the AI service providers are benefiting significantly from the use cases you bring forward for trial as well as the vast amounts of data being processed in their platforms. These points should be leveraged to negotiate a better deal.
Identify business risk cycles & inherent bias
As with any implementation, it is important to assess the various risks involved. As technologies become increasingly interconnected, entry points for potential data breaches and risk of potential compliance claims from indirect use also increase. What security measures are in place to protect your data and prevent breaches? How will indirect use be measured and enforced from a compliance standpoint? Another risk AI is subject to is unintentional bias from developers and the data being used to train the technology. Unlike traditional systems built on specific logic rules, AI systems deal with statistical truths rather than literal truths. This can make it extremely difficult to prove with complete certainty that the system will work in all cases as expected.
Develop a sourcing and negotiation plan
Using what you gained in the first three steps, develop a sourcing and negotiation plan that focuses on transparency and clearly defined accountability. You should seek to build an agreement that aligns both your enterprise’s and service provider’s roadmaps and addresses data ownership and overall business and security related risks. For the development of AI , the transparency of the algorithm used for AI purposes is essential so that unintended bias can be addressed. Moreover, it is appropriate that these systems are subjected to extensive testing based on appropriate data sets as such systems need to be “trained” to gain equivalence to human decision making. Gaining upfront and ongoing visibility into how the systems will be trained and tested will help you hold the AI provider accountable for potential mishaps resulting from their own erroneous data and help ensure the technology is working as planned.
Develop a deep understanding of your data, IP, commercial aspects
Another major issue with AI is the intellectual property of the data integrated and generated by an AI product. For an artificial intelligence system to become effective, enterprises would likely have to supply an enormous quantity of data and invest considerable human and financial resources to guide its learning. Does the service provider of the artificial intelligence system acquire any rights to such data? Can it use what its artificial intelligence system learned in one company’s use case to benefit its other customers? In extreme cases, this could mean that the experience acquired by a system in one company could benefit its competitors. If AI is powering your business and product, or if you start to sell a product using AI insights, what commercial protections should you have in place?
In the end , do realize the enormous value of your data, participate in AI readiness, maturity workshops and immersion sessions and identification of new and practical AI use cases. All of this is hugely beneficial to the service provider’s success as well and will enable you to strategically source and win the right AI deal.
(AIQRATE advisory & consulting is a bespoke global AI advisory & consulting firm and provides strategic advisory services to boards, CXOs, senior leaders to curate , design building blocks of AI strategy , embed AI@scale interventions & create AI powered enterprises . Visit www.aiqrate.ai , reach out to us at email@example.com )
AI for Strategic Innovation
The extra ordinary promise of AI : Global & Indian enterprises have a lot to gain from unleashing innovation with AI —but harnessing their potential demands focused investment and a new way of working with external partners.
Here are few salient features of how AI has become game changing trend in spurring innovation; existing challenges and few strategic approaches of unlocking innovation with AI :
- 22% growth : From 2015 through 2019, disclosed private investment in seven deep tech sectors grew an average of 22% per year, equaling nearly $60 billion in total investment. Corporate venture capital is also playing an increasingly active role.
- Total investment : Nearly $60 Billion Invested in Deep Tech’s Fastest-Growing Sectors in 2019; Artificial intelligence corners close to $25 Bn
- About 1800 AI led startups in the US accounted for roughly half of this total investment, but other countries are catching up fast.
- Complex ecosystems : Multiple types of players including startups, venture capital firms, governments, universities and research centers, and early-adopter user groups
- Dynamic Interactions : Few central orchestrators; business relationships based on informal networks rather than formal contracts
Strategic approaches of unlocking innovation with AI :
- Cooperate in order to compete : Think beyond the enterprise’s immediate goals; commit to a long-term vision for the development of the ecosystem as whole
- Identify capabilities that add value : Define what the enterprise can offer to nurture the ecosystem and bring AI to market—not only money but also access to customers, data, networks, mentors, and technical experts
- Don’t pick winners in advance : AI startups are evolving rapidly. Continuously monitor the ecosystem to identify successful startups, applications, and business models as they emerge
- Blur the boundaries with partners : Make it easy for AI partners to navigate your corporate system. Define a clear role for them in your innovation strategy, ensure senior-executive sponsorship, and engage the core businesses
- Streamline decision making and governance : Success requires partnering more nimbly with fast-moving AI startups. Embrace agile ways of working.
- Develop breakthrough solutions by combining expertise from previously unconnected fields or industries. Be alert for game hanging opportunities that deliver both economic and social value.
AI will transform business and society in the future. The time to craft a AI strategy for unleashing innovation is now.
AIQRATE works closely with global & Indian enterprises , GCCs , VC/PE firms and has an extensive yet curated database of 1000 + global AI startups , boutique and niche firms benchmarked on our “Glow Curve” assessment.
(AIQRATE advisory & consulting is a bespoke AI advisory and consulting firm and provide strategic advisory services to boards , CXOs, senior leaders to curate , design building blocks of AI strategy , embed AI@scale interventions and create AI powered enterprises . visit : www.aiqrate.ai ; reach out to us at firstname.lastname@example.org )
AI led strategy for business transformation : A guided approach for CXOs
Business transformation programs have long focused on productivity enhancements —taking a “better, faster, cheaper” approach to how the enterprise works. And for good reason: disciplined efforts can boost productivity as well as accountability, transparency, execution, and the pace of decision making. When it comes to delivering fast results to the bottom line, it’s a proven recipe that works.
The problem is, it’s no longer enough. Artificial Intelligence enabled disruption are upending industry after industry, pressuring incumbent companies not only to scratch out stronger financial returns but also to remake who and what they are as enterprises.
Doing the first is hard enough. Tackling the second—changing what your company is and does—requires understanding where the value is shifting in your industry (and in others), spotting opportunities in the inflection points, and taking purposeful actions to seize them. The prospect of doing both jobs at once is sobering.
How realistic is it to think your company can pull it off? The good news is that AIQRATE can demonstrate that it’s entirely possible for organizations to ramp up their bottom-line performance even as they secure game-changing portfolio wins that redefine what a company is and does. What’s more, AL led transformations that focus on the organization’s performance and portfolio appear to load the dice in favor of transformation results. By developing these two complementary sets of muscles, companies can aspire to flex them in a coordinated way, using performance improvements to carry them to the next set of portfolio moves, which in turn creates momentum propelling the company to the next level.
Strategic Steps towards AI led Transformation:
This aspect covers AI led “portfolio-related” moves. The first is active resource reallocation towards building AI led transformation units, which I define as the company shifting more than 20 percent of its capital spending across its businesses or markets over ten years. Such firms create 50 percent more value than counterparts that shift resources at a slower clip.
Meanwhile, a big move in programmatic M&A driven by AI led spot trending—the type of deal making that produces more reliable performance boosts than any other—requires the company to execute at least one deal per year, cumulatively amounting to more than 30 percent of a company’s market capitalization over ten years, and with no single deal being more than 30 percent of its market capitalization.
Making big moves tends to reduce the risk profile and adds more upside than downside. The way I explain this to senior executives is that when you’re parked on the side of a volcano, staying put is your riskiest move.
AI led Transformations that go ‘all in’ by addressing both a company’s performance and its portfolio yield the highest odds.
The implication of these transformation stories is clear: approaches that go all in by addressing both a company’s performance and its portfolio yield the highest odds of lasting improvement. Over the course of a decade, companies that followed this path nearly tripled their likelihood of reaching the top quin tile of the AI transformation power curve relative to the average company in the middle.
Play to win with AI
Life would be simpler if story ended here. However, you’re not operating in a competitive vacuum. As I described earlier, other forces influence your odds of success in significant ways—in particular, how your industry is performing. Research studies have indicated that companies facing competitive headwinds would face longer odds of success than those with tailwinds.
Companies that combined big performance moves with big portfolio moves (including capital expenditures, when not the only portfolio move employed) saw a big lift in their odds. Life is still challenging for these companies—their net odds are dead even—yet this is superior to the negative odds of the other situations.
Winning thru competitive advantage with AI
In an improving industry, the returns to performance improvement are amplified massively. This runs contrary to the very human tendency of equating performance transformations with turnaround cases
The takeaway from all this is that two big rules stand out as commonly and powerfully true whatever your context: first, get moving with AI , don’t be static; second, go all in if you can with AI led transformation programs —it’s always the best outcome (and also the rarest).
Running the AI led transformation program
In my experience, the companies that are most successful at transforming themselves with AI ,sequence their moves so that the rapid lift of performance improvement provides oxygen and confidence for big moves in M&A, capital investment, and resource reallocation. And when the right portfolio moves aren’t immediately available or aren’t clear, the improved performance helps buy a company time until the strategy can catch up.
To illustrate this point, consider the anecdote about Apple that Professor Richard Rumelt describes in his book, Good Strategy/Bad Strategy. It was the late 1990s; Steve Jobs had returned to Apple and cleaned house through productivity-improving cutbacks and a radically simplified product line. Apple was much stronger, yet it remained a niche player in its industry. When Rumelt asked Jobs how he planned to address this fact, Jobs just smiled and said, ‘I am going to wait for the next big thing.’
While no one can guarantee that your “next big thing” will be an iPod-size breakthrough, there’s nothing stopping you from laying the groundwork for a successful AI led transformation. To see how prepared, you are for such an undertaking, ask yourself—and your team—the following five questions. I sincerely hope they provoke productive and transformative discussion among your team.
1.Where is the new business value chain that’s driven by AI
Achieving success with big, portfolio-related moves requires understanding where the business value flows in your business and why. The structural attractiveness of markets, and your position in them, can and does change over time. Ignore this and you might be shifting deck chairs on the Titanic. Meanwhile, to put this thinking into action, you must also view the company as an ever-changing portfolio. This represents a sea change for managers who are used to plodding, once-a-year strategy sessions that are more focused on “getting to yes” and on protecting turf than on debating real alternatives. Get high-powered decision-making algorithms to navigate you thru this transformation.
2. Put your money in building an AI led strategy
Only 10% of the US fortune 200 companies have AI led strategy; this is an impending strategic aspect that cannot be ignored. The dimensions of reimagining customer experience, building innovative products and services and transforming the businesses need to have an AI led strategy move by the CXOs
3.Are you ready for disruption?
Increasingly, incumbent organizations are getting to the pointy end of disruption, where they must accelerate the transition from legacy business models to new ones and even allow potentially cannibalizing businesses to flourish. Sometimes this requires a very deliberate two-speed approach where legacy assets are managed for cash while new businesses are nurtured for growth.
4.Will our company take this seriously?
Embracing AI led transformative change requires commitment, and gaining commitment requires a compelling change story that everyone in the company can embrace. Philips recognized this in 2011 when it launched its “Accelerate” program. Along with productivity improvements and portfolio changes (including a big pivot from electronics to health tech), the company shaped its change story around improving three billion lives annually by 2030, as part of a broader goal of making the world healthier and more sustainable through innovation. Massive thrust and investment was laid by Phillips leadership team on AI led transformation programs.
5.Is the leadership ready for the transformation?
Leading a successful AI led transformation requires a lot more than just picking the right moves and seeing them through. Among your other priorities: build momentum, engage your workforce, and make the change personal for yourself and your company. All of this means developing new leadership skills and ways of working, while embracing a level of commitment as a leader that may be unprecedented for you.
In the end, AI led strategy for transformation is a process and start of a journey …. embrace it or feel the heat of leaving behind. The new age competition is agile and nimble and AI led transformation strategy is a right move to thwart the competition.