Part 1. Machines Working Faster

Everything that can be automated will be...

The acceleration of machine speeds already has profound and far-reaching impacts on computational power, data processing, communication networks, automation, healthcare, scientific research, and consumer technology. This trend is reflected by a surge in the marketplace valuations of companies leading the AI revolution, as investors explore their potential for explosive growth. Coupled with the prospect of precision healthcare, renewable energy storage, smart grids, digital twins, solar desalination, and many other innovations, frontier technology is not only consolidating its position as a commercial and economic game-changer but also demonstrating its potential to combat some of humanity’s most pressing problems.

Two decades ago, software engineering expert Watts S. Humphreys argued that every organization, regardless of industry, product, or service, was becoming a technology company. This is truer than ever today, as frontier technologies are fueling new growth models based on their potential for upsurges in productivity and ground-breaking innovations. Virtually every firm mired in earlier industrial age practices and technologies risks being disrupted by competitors deploying ever more powerful digital capabilities.

For business leaders, adapting and making decisions at the speed of digitalization is a new challenge. Here are a few ideas to help adjust to this new pace:

• Aligning to digital’s combinatorial and exponential progression
• Developing Use Cases with keen insights into organizational and people readiness
• Expecting the unexpected – digital technologies are emergent
• Managing productivity expectations, and avoiding hype and costly missteps.

Aligning to digital’s combinatorial and exponential progression

Frontier tech’s movement towards new value creation is largely driven by linear-defying speed of progression, compounded by the convergence of technologies (such as drones interacting with IoT) and the network effect of learning at scale, that in combination enable innovations to interact even faster with the business world and accelerate the ongoing 4th Industrial Revolution.

The creation of synthetic data is significantly accelerating digital advancements, development, and deployment. Here, generative AI models train on real work data samples first, and once the algorithms learn the patterns, correlations, and statistical properties of the sample data, the generator can create statistically identical, synthetic data. This solves issues of data scarcity, quality, privacy, and testing. It also enables more robust, scalable, and innovative solutions. This acceleration has the potential to impact various sectors, from healthcare and finance to autonomous systems and beyond, driving developments in AI technology and its applications.

The “time travel” or chronological compression attribute of the digital revolution has also produced truly new phenomena (not just simply more powerful or efficient versions of things past), that were previously beyond our human conception or perception.

Knowing these important attributes of digital, leaders can be both strategic and forward-looking by envisioning and planning for multiple futures using flexible approaches based on optionality. Failing fast, learning quickly on multiple timescales, and maintaining momentum are all “working at pace” capabilities and strategies where leaders embrace uncertainty and find opportunities to profit amid the turbulence of today’s business environment.

Developing Use Cases with keen insights into organizational and people readiness

For business participants and their partnerships with internal IT, it may be helpful to distinguish between the foundational capabilities needed to support digital at scale - typically within the realm of IT - from the prioritization of business use cases for its application. Use cases provide the critical context needed to move from a discussion of data, content, process, and transactions to envisioning the future operating model “in action”, with new human and machine knowledge paradigms and organizational capabilities. Use cases are particularly important for augmented human-machine work as they define new work design, including expanding relationships, exchanges and flows, dependencies, and boundaries. These also provide the basis for considering organizational readiness and impact on people-process-policies, even culture, and risk. It’s important not to miss the subtlety here – augmented work design is an opportunity for reimagining work, not just automating it. This requires transformational thinking by business leaders, who need to build for exponential growth and anticipate the impact of change on people while remaining focused on organizational purpose. These are the critical business decisions required to maximize the value of digital in new ways of working, and an important foundation of the business-IT relationship.

Asking the right questions here requires a minimal viable appreciation of digital’s capabilities and its potential impact:

• Which tasks fit where on the spectrum of fully automated vs human-led (and AI-enhanced) processes, workflow, and decisioning?
• How can the organization ensure seamless, end-to-end human-AI collaboration?
• What will humans need to know about machine logic to collaborate?
• Are there ethical issues that require consideration? Issues that challenge our values, commitments, and/or purpose as an organization?
• What are the implications of various models on workforce planning and development? 

By addressing questions such as these, business leaders bring the critical context, insights, and prioritizations for integrating digital into the organization and creating the “glue” between a “shiny objects” view of technology, and the clarity of the to-be business roadmap and operating model. Additionally, businesses must own and advocate for the priority of organizational and people readiness including policy, communications and change, and governance.

Expect the unexpected – digital technologies are emergent

Digital's dynamism and capacity for unexpected actions distinguish it from prior technologies in several significant ways. Traditional deployments typically follow predefined rules and procedures, whereas digital/AI systems, especially those employing machine learning and deep learning, exhibit more complex and adaptive behaviors, such as the ability to improve their performance over time without explicit reprogramming, or making instantaneous decisions based on patterns and insights derived from data. 

Unlike traditional technologies, AI can also exhibit unexpected behaviors, especially in complex systems or when dealing with incomplete or evolving datasets, or even develop new strategies or solutions that were not explicitly programmed. AI’s capacity for autonomy (e.g., operating self-driving cars) and making unilateral decisions based on data analysis, pattern recognition, and probabilistic reasoning without explicit human intervention, raises the opportunity for ground-breaking discovery, even while posing questions regarding control, accountability, and bias.

Managing productivity expectations, and avoiding hype and costly missteps
Faster, less people-intensive technologies set high bars of expectation for significant work productivity gains, often with headcount reductions. Success in these transformational strategies, though, requires governance and oversight in risk management, organizational design, methodology, and program management - more so than in product/tech selection.

Here are some typical shortcomings:

1. Hype and Marketing, including exaggerated claims (by vendors) or misinterpreting/not understanding technology claims (by internal staff) about the immediate benefits and capabilities of new technologies, leading to inflated expectations.
2. Underestimating the complexity, effort, training and/or adaptation needed for seamless integration and interoperability.
3. Over-engineering business complexity by not distinguishing between necessary and needed intricacy vs seizing the opportunity to simplify, reduce, or eliminate organizational clutter.
4. Undervaluing the significance of current state technical debt relative to future state innovation.
5. Launch of new technology at scale or globally, with insufficient pilot/testing, and/or pilots don’t accurately reflect the complexities of real-world use cases.
6. Organizational readiness is ignored or underestimated, overlooking the “system” holistically and missing supporting infrastructure, related systems rationalization, and/or not redesigning policy and processes sufficiently to realize desired results.
7. Underestimating costs/challenges and failing to account for total cost of ownership including migration, integration, governance, and sustainability.
8. Unrealistic timelines that are either missed or, more critically, met but with post-launch fundamental flaws, serious experience problems, and/or overall instability.
9. Underestimating data/content (and related processes) evolution and their migration challenges.
10. Overestimating early productivity gains, resulting in premature staff reductions and/or overwhelming existing staff with more work than anticipated.

These are fundamental issues, and as organizations accelerate this relationship journey with frontier technologies, the evolution of work will necessarily take shape over time. It’s best to think we’re in the early phases of a relationship with cognitive machines that is both unpredictable and fraught with hidden challenges while balancing a desire to envision the future while delivering on the core business today.