I have used this in some board and executive briefings as a discussion starter, and first shared this publicly yesterday at Agentforce World Tour Sydney in the opening keynote Executive Experience panel.

Worker Value Shifts

At the threshold of organizational redesign, we see three primary mechanisms through which AI agents create value:

Automation represents the substitution of specific human functions with digital workers. This direct replacement pattern forms the foundation of many initial AI implementations.

Productivity enhances human effectiveness through strategic task allocation between human and digital workers. This is not merely about speed, but about optimizing who (or what) performs each task based on comparative advantages.

Augmentation provides cognitive support that fundamentally expands human capabilities. Rather than replacing or dividing labor, this pattern creates new possibilities through combined human-AI synergies.

Elemental Humans + AI Workflows

In contrast to traditional sequential workflows—whether performed entirely by humans or through basic process automation—two core collaborative patterns are emerging:

Human-AI Sandwich structures work into three distinct phases: humans frame tasks by providing intent and context; AI performs the execution based on these parameters; and humans review and refine the output, ensuring quality and alignment with objectives.

AI with Humans-in-the-Loop inverts this relationship. Here, the AI system performs the primary task or decision, humans provide approval or feedback at critical junctures, and the AI executes or improves based on this input.

These elemental workflows represent a significant departure from linear processes, creating feedback loops that leverage the strengths of both human and artificial intelligence.

High-Performance Humans + AI Workflows

The essence of truly transformative organizational redesign lies in dynamic work orchestration. This approach moves beyond static, predefined workflows to create adaptable systems where:

• Human and digital workers collaborate in outcome-oriented processes
• Workflows reconfigure in response to changing conditions
• Orchestration agents, under human supervision, continuously optimize these evolving systems

This dynamic orchestration represents the frontier of organizational design—not simply automating existing processes, but fundamentally reimagining how work gets done through fluid human-AI collaboration.

In essence, successful organizational redesign will not be about replacing humans with AI, but about creating sophisticated collaborative frameworks where each contributes their unique strengths. The patterns outlined in this framework provide a starting point for organizations seeking to navigate this transformative threshold.

We are very early in discovering the structures that work best, and how to configure across multiple dimensions, including internal value creation based on humans + AI, a diverse array of enhanced customer offers, ecosystemic value co-creation, model scalability, and the coming rise of AI agents as customer intermediaries.

The framework below distills the most important frames for understanding how business models are shifting. This is just an initial top-level view of the elements, I will be sharing more detail on many aspects of this in the near future.

See full size framework

AI’s Impact on Business Models and Value Creation

Driving Forces of AI Evolution

• Continued exponential growth in AI capabilities
• Decreasing cost of AI and predictions
• Increasing value of data ecosystems
• Automation of cognitive tasks
• Real-time decision-making
• Acceleration of the economy

Scalable Efficiency:

• Automation and productivity boost
• Accelerating decision cycles
• Resource optimization at scale
• Continuous process iteration

Enhanced Value Propositions:

• AI-augmented services/products
• Smart product offerings
• Superior customer experience
• Proactive, anticipatory solutions

Customer Value Co-Creation:

• Hyper-personalized services and products
• Continuous adaptive customer value creation
• Enhancing customer decision-making
• Real-time iterative customization

AI as Customer:

• M2M market evolution
• AI-driven decision-making
• Autonomous customer interactions
• Agent-to-agent transactions

Restructuring Value Delivery:

• Dynamic human-AI collaboration
• Hybrid decision-making workflows
• Agile service delivery models
• Humans + AI operational balance

Shifts in Platforms and Ecosystems:

• Data as core asset
• Intelligent platform strategy
• Ecosystem-driven innovation
• Open-source AI collaboration

Evolving Your Business Models:

• AI-first organizational design
• Innovation through transformation
• Capability for rapid scaling
• Continuous model adaptation

As part of my session I shared this framework on unbundling and rebundling higher education. I explain and expand below the image.

LLMs are most effectively applied at well-defined points within broader processes, using precise prompts and context for highly specific tasks such as content generation, ideation, structure mapping, analysis, adaptation, synthesis, evaluation.

The first step is to pull back to the broader process and identify where AI can add value to complex cognitive and creative processes, as shown in this chart. Applications of the chart are shared below.


Click on image for full size pdf

The chart illustrates that we can break down creative processes into a series (not necessarily linear) of specific activities.

Examining what is common and different for creative processes across domains can help cross-pollinate approaches and ideas.

Each stage in the highly simplified processes in the chart can be broken down into more specific elements.

Clarifying specific activities within creative processes makes them far more amenable to AI augmentation.

A range of AI techniques can be applied to complement human thinking across each stage and many of their sub-elements.

Part of the point of the diagram is to highlight that we should not be trying to get AI to do entire creative processes. We should be applying AI to specific Humans + AI tasks within the processes.

We are early in unpacking creative processes and augmenting each stage with AI. There’s immense value creation to come on this journey.

Image: Midjourney

By the very nature of discovery, we cannot know what we will learn. The coming advances in life and brain sciences and medicine alone will be absolutely transformative, let alone those in energy, materials science, food, and more.

This framework is a high-level distillation of a few of the cognitive tools and techniques for GenAI to assist scientists. As always it is a Beta version 1, please let me know any feedback or input to improve it!


Click on the image for full size pdf

Generative AI in Scientific Discovery

Observation and Curiosity

Example application: Generative Summaries: AI creating summaries and highlighting key findings from large datasets or experimental results to prompt new questions.
In practice: Analyzing oceanographic data to highlight unusual patterns in marine life migration, suggesting new areas of study.

Literature Review

Example application: Contextual Analysis: AI generating context-aware reviews by linking new research to existing literature, helping scientists quickly understand the relevance.
In practice: Producing a comprehensive summary of recent advancements in CRISPR technology, tailored to a researcher’s specific focus area. patterns in marine life migration, suggesting new areas of study.

Idea Generation

Example application: Cross-Disciplinary Connections: AI generating potential interdisciplinary research ideas by combining concepts from different fields.
In practice: Proposing innovative uses of nanomaterials in drug delivery by integrating knowledge from materials science and pharmacology.

Hypothesis Formulation

Example application: Hypothesis Suggestion Engines: Generative AI proposing new hypotheses by identifying gaps and connections in current research.
In practice: Suggesting hypotheses about the impact of microplastics on marine ecosystems based on multi-factor analysis.

Experiment Design

Example application: Simulation-Based Design: AI generating virtual experiments to predict outcomes and refine real-world experimental designs.
In practice: Developing a detailed protocol for a synthetic biology experiment, including optimal conditions and potential pitfalls.

Data Collection

Example application: Enhanced Instrumentation Control: AI generating instructions for automated control of laboratory instruments to optimize data collection.
In practice: Developing customized survey questions for a large-scale sociological study on urban migration patterns.

Data Analysis

Example application: Generative Data Interpretation: AI generating interpretations of complex datasets, highlighting unseen emergent patterns and trends.
In practice: Analyzing and interpreting large genomic datasets to surface potential gene-disease associations.

Interpretation

Example application: Contextual Explanation: Generative AI providing context-aware explanations and interpretations of research findings.
In practice: Generating interactive visual explanations of climate model data, helping researchers and policymakers understand potential impacts.

Writing and Communication

Example application: Language Enhancement: AI improving the clarity and readability of scientific writing, translating complex ideas into accessible language.
In practice: Writing grant proposals that align with funding agency guidelines and highlighting novelty and impact.

Peer Review

Example application: Bias Detection and Suggestions: AI identifying potential biases or gaps in research and suggesting areas for improvement or additional study.
In practice: Summarizing the main points from peer reviews of a submitted manuscript, providing authors with clear and actionable feedback.

What this debate really boils down to is the business models for news media.

A key point beyond the extinction or survival of the industry is its diversity, especially in whether it can extend beyond a handful of major conglomerates and mastheads globally.

Subscriptions have finally proven to be a sustainable model for some major players, plus an increasing range of niche players, often at a micro-scale. Yet they are largely not working for hyperlocal and many mid-sized news brands.

The recent News Corp – OpenAI deal for potentially $250m+ over five years sounds promising, but likely few media organizations will be able to cut these deals.

Media related to buying decisions in verticals such as travel, healthcare, technology, real estate, finance, automotive, fashion will always have associated revenue streams. However these may not directly support quality news.

In a world of misinformation people want a source of truth. In whatever way this is provided, this is in essence news media. If people value that, which they increasingly will, then there will be business models around it.

It has long struck me that media provides essential input to almost every decision we make, ranging across how we vote, where we live, the jobs we get, the investments we make, and how we lead teams and organizations in changing times.

The value in making better decisions is immense. Media companies have an opportunity to go further, assisting better decisions rather than simply providing inputs.

The reality is that we can’t anticipate the successful media business models of the future. It will take experimentation to discover what will work in a world in which AI is swiftly changing the landscape.

However I believe that the many brilliant people in the media industry will uncover the models that will prove to be successful, providing high-value trustworthy news and information to us all long into the future.  

The other major Humans + AI application I’ve been working on is in institutional investment decision-making. I am currently shifting more of my attention and work into this space. 

This framework provides a very high-level slice of a few of the many human and generative AI roles in different phases of portfolio decision-making.

Click on the image for full-size pdf

The activities indicated across the human and GenAI phases are illustrative only. There are a wide range of other potential roles.

As in other Humans + AI frameworks I’ve shared, each of these roles can and should involve both humans and AI.

Here is a brief description of the highlighted GenAI roles across the investment phases.

Investment Parameters

Suggest alternative frames: Proposes different options for investment parameters based on objectives and hard constraints.
Sustainability gap analysis: Analyzes where current parameters may negatively impact sustainability objectives and alternatives.

Investment Thesis

Identify emergent trends: Surface trends in consumer behavior, industry sectors, and macro economy that could impact investment decisions.
Test robustness: Identify across a range of scenarios where proposed investment hypotheses would be resilient or deficient.

Asset Allocation

Model alternatives: Examine potential outcomes for a range of alternatives to asset allocation using qualitative parameters.
Identify novel risk scenarios: Propose new, unforeseen, or multi-factor risk scenarios that could affect asset allocation.

Sector Analysis

Simulate industry evolution: Provide a variety of scenarios for the industry evolution, competitive landscapes, and profitability.
Challenge analyses: Critical evaluation of sector forecasts by identifying complementary perspectives to enrich models.

Investment Selection

Predictive analytics: Predictive analyses using domain-unbounded superforecasting methodologies to complete machine learning models.
Filter and surface options: Swiftly assess and propose specific assets to match portfolio strategies while highlighting critical factors.

Due Diligence

Quantitative analysis: Perform quantitative analyses to assess the financial health of potential investments.
Identify non-obvious risks: Gen AI identifies subtle and non-obvious risks that might be overlooked in traditional due diligence.

Decision

Red-teaming: Red-teaming exercises to identify key assumptions and provide incisive counterpoints to investment proposals.
What-if analyses: Identify relevant scenarios and model decisions against multi-dimensional variations in investment context.

Coming from a capital markets background, I believe there is little more valuable than improving the allocation of capital for superior financial and social outcomes.

The considered application of GenAI for a Humans + AI investment process probably has the greatest potential for positive impact in this space.  

I will be sharing more of my work in this space as there are frameworks and insights that can be shared publicly.

The story recounts how a contemporary writer, Pierre Menard, rewrites Cervantes’ epic Don Quixote word for word. However he writes it entirely from his own context, making the text laden with layers of meaning missing from the original. The reviewer played by Borges in the piece finds the new Quixote, coinciding word for word with the original, enriched, astounding, and more significant than the original.

If a human writer and AI create exactly the same text, which has more value?

Every word written by a person can only come from their complete context and life experience. Any writing utterly reflects the writer, however twisted it might be by education or pretention. AI has no context. It is simply generating a plausible series of words. 

While not everyone has Borges’ sensibility and attunement to the state of the writer, I think most people reading a marvelous story wonder who it was who wrote those words, how they came to the imagination that produced this work. They are engaged not just with the output, but intuiting and relating to the person created it.

Which brings us to the question of how people will treat AI-created versus human-created art. 

I believe that, save perhaps for formulaic thrillers and romances, people will always prefer human-created over AI-created art. They can feel connected to the author and their experience, if they know there is one, but there can be no relationship with an AI that ingests and spits out words.

Then again, perhaps AI, in its statistical recombination of billions of words produced by humans, is distilling an essence of humanity. We are relating not to the machine, but to an amalgam of human experience that we can deeply identify with.

Certainly, Borges’ story highlights that the experience of creativity is just as much about who created it as the creation. 

Link to the story: Pierre Menard, Author of The Quixote.

In this world AI wil be transformative, in most cases not by supplanting humans in the decision-making process, but in playing a role in ‘Humans + AI’ decision-making.

I was recently interviewed for the NAB Digital Next podcast for an episode titled Futurist Ross Dawson on humans and AI achieving more together. You can listen here, with some distilled reflections below.

The conversation was far-ranging in explaining the concept of Humans + AI and its implications in finance and large organizations.

One segment focused on various aspects of financial decision-making. Here is a summary of the main points I shared on the evolving role of AI in consumer, portfolio and organizational decisions:

Consumer financial decisions

Personalization and Empowerment
AI enables the distillation of complex financial information into personalized advice tailored to individuals’ learning styles and emotional states. This personalized approach empowers consumers to make informed decisions aligned with their unique circumstances and preferences.

Enhanced Financial Literacy
Technology can bridge the gap in financial literacy by externalizing consumers’ approaches to financial management and guiding them through their financial positions. This assistance is particularly valuable for individuals lacking financial literacy, offering support and guidance in navigating financial complexities.

Value Creation for Consumers
The ultimate goal is to provide services that help consumers feel in control of their financial destinies. By harnessing AI to analyze consumer data and provide personalized insights, financial institutions can offer valuable tools and guidance that empower individuals to make better financial decisions.

Portfolio decision-making

Complex Decision-Making Process
Portfolio decision-making involves analyzing investment basics, conducting stress tests, and exploring various scenarios. This process is multifaceted and requires careful consideration of factors such as asset allocation, risk management, and strategy development.

AI-Assisted Decision-Making
AI can play a crucial role in assisting investors by providing recommendations with justifications. By leveraging AI algorithms, investors can make informed decisions while maintaining control over their portfolios. AI helps navigate the complexities of investment by distilling vast amounts of data into actionable insights.

Risk Management and Scenario Analysis
AI facilitates risk management by conducting stress tests and scenario analysis. By simulating different market conditions and potential outcomes, investors can better understand the risks associated with their portfolios and make adjustments accordingly.

Role of AI in Institutional Settings
Within financial institutions, AI serves as a valuable tool for portfolio managers and investment professionals. It complements human expertise by providing data-driven insights and facilitating decision-making processes. AI’s role is particularly prominent in tasks that require cognitive capabilities and domain knowledge.

Organizational Decision-Making and Leadership

Trust and Communication
Effective leadership involves fostering trust between employees and leadership. Clear communication about the organization’s goals and intentions regarding AI integration is crucial for gaining employee buy-in and minimizing resistance to change.

AI Integration Strategy
Financial institutions must view AI as more than just a tool for efficiency but as a means to reimagine workflows and optimize human capabilities. Leaders should develop a strategy for integrating AI into organizational processes while emphasizing its role in augmenting human decision-making rather than replacing it.

Reconfiguration of Workflows
AI enables the reconfiguration of workflows to optimize human-AI collaboration. By creating modularized, configurable workflows, organizations can allocate tasks dynamically and capitalize on the unique strengths of both humans and AI.

Empowering Employees
Leadership should empower employees to experiment with AI tools and develop competencies in using them effectively. Encouraging experimentation and providing support for learning and development can accelerate AI adoption and drive innovation within the organization.

Consideration of Stakeholders
Financial institutions must consider the impact of AI on all stakeholders, including customers, employees, shareholders, and the broader community. Integrating AI should focus on augmenting human capabilities and creating positive outcomes for all stakeholders.

I endlessly read people arguing that humans will be to AI as animals or insects are to humans. They envision a future where AI’s relentless advancement transcends every faculty we possess.

The countervailing stance is that human potential is unlimited. We have deliberately and consistently increased our capabilities and knowledge, and now we will use the tools we have created to continue to advance.

The rise of AI has intensified this debate, leading us to question: Are we, as humans, inherently limited or unlimited?

My thinking on this was clarified and crystalized by reading David Deutsch’s seminal work “The Beginning of Infinity”, in which he lays out an extensive and powerful case for human knowledge and abilities being unbounded.

He argues that humans are “universal explainers” who have created and learned to use scientific principles to indefinitely improve our theories, knowledge, and understanding,

At any point our understanding is limited and incorrect, but by continuing to apply the same principles we will consistently and indefinitely advance our knowledge.

He writes:

The astrophysicist Martin Rees has speculated that somewhere in the universe ‘there could be life and intelligence out there in forms we can’t conceive. Just as a chimpanzee can’t understand quantum theory, it could be there are aspects of reality that are beyond the capacity of our brains.’ But that cannot be so. For if the ‘capacity’ in question is mere computational speed and amount of memory, then we can understand the aspects in question with the help of computers – just as we have understood the world for centuries with the help of pencil and paper. As Einstein remarked, ‘My pencil and I are more clever than I.’

In terms of computational repertoire, our computers – and brains – are already universal (see Chapter 6). But if the claim is that we may be qualitatively unable to understand what some other forms of intelligence can – if our disability cannot be remedied by mere automation – then this is just another claim that the world is not explicable. Indeed, it is tantamount to an appeal to the supernatural, with all the arbitrariness that is inherent in such appeals, for if we wanted to incorporate into our world view an imaginary realm explicable only to superhumans, we need never have bothered to abandon the myths of Persephone and her fellow deities.

So human reach is essentially the same as the reach of explanatory knowledge itself.

The track record of humanity, which in just the last few thousand years has grown from superstitions to understanding in great depth our Universe from sub-atomic particles to the structure of the cosmos, sitting on a planet in a far-flung galaxy, is testament to our ability to develop knowledge.

Indeed, the pace of (human) scientific progress and knowledge is not just fast, it is accelerating, by just about any measure you choose.

People compare the exponential technologies underlying AI with our finite cognition and conclude we will be left behind. This is so deeply flawed that I will address this in more detail in another post.

In short, humans are very clearly not static. We co-evolve with the technologies we have created, constantly extending the boundaries of what it means to be human.

It is possible that to keep pace we will need to augment ourselves with brain-computer interfaces and other cognitive amplification technologies. As I wrote in my 2002 book Living Networks,

the real issue is not whether humans will be replaced by machines, because at the same time as computing technology is progressing, people are merging with machines. If machines take over the world, we will be those machines.

Believing we’re doomed to be dwarfed by super-intelligent AI is essentially betting against humanity.

It is giving up. It is lacking faith in humanity. It’s a surrender to superstition, to the idea that there is something undefined and unknowable beyond our capacity to imagine or understand.

The essence of being human is this: We face and solve problems and we progress. The concept of things we cannot understand goes against everything that humans have demonstrated themselves to be.

In an era defined by the rapid rise of AI, it is crucial that we maintain faith in the human capacity for limitless growth and expansion. The unfolding story of our species is not one of succumbing to imagined limits but one of constantly redefining what is possible.

Our future is inevitably one of Humans + AI – our species amplified by the intelligences we have created.

We will not be subsidiary players in this union. Our ability to understand and grow and frame what this incredible pairing can achieve is unlimited.

At this point we have no solid evidence how this will turn out. It is your choice:

Believe humans are intrinsically limited and that we will be as cockroaches to superior intelligences.

Or bet on a species that is intelligent and adaptable enough to have created everything we have so far, and our ability to continue to progress and grow, harnessing the power of our inventions.