[David’s Note]

For over a decade, the “smart city” has been a buzzword in urban development, yet in practice, it often merely amounts to “data visualisation”. Behind those visually impressive data dashboards, day-to-day operations still rely on inefficient human approvals, inter-departmental buck-passing, and protracted budget cycles. This week, we bring you our curated selection of McKinsey & Company’s latest heavyweight, cutting-edge report: How AI-native public infrastructure changes how cities operate.

The report hits the nail on the head: the true revolution lies not in the sheer number of added Internet of Things (IoT) devices, but in where “intelligence” is situated within the system. As urban infrastructure crosses into the “AI-native” era, the city will cease to be a sluggish bureaucracy. Instead, it will transform into a colossal, distributed computing platform, equipped with millisecond-level autonomous, closed-loop capabilities of “perception, decision, and execution”. This article will fundamentally overturn your traditional understanding of urban governance.

▍ Core Insights & Key Takeaways at a Glance

1. A Core Paradigm Shift: From “Observation and Reporting” to “Decision and Execution”

Traditional smart city technology stops at the “observation” level: systems identify problems, and humans solve them. In contrast, the “AI-native city” achieves a fundamental structural leap—within explicit policy and regulatory guardrails, the system begins to autonomously make decisions and execute them. The urban control loop shifts from monthly reviews to “millisecond-level” continuous execution. The city begins to behave much like modern software: versioned, observable, testable, and capable of autonomous operation.

2. Redefining the Five Key Characteristics of an AI-Native City

• Operating like a distributed computing system: Bidding farewell to rigid, siloed, monolithic systems. Subsystems like transport, power, and water management operate independently yet coordinate seamlessly, much like “microservices” in a cloud architecture. Localised failures (e.g., congestion on a specific road) are automatically isolated and dynamically rerouted by the system, rather than triggering city-wide paralysis.
• High-resolution sensing as the “urban nervous system”: Shifting from periodic manual inspections to high-density continuous perception via edge computing. The system can proactively identify stress signals in physical assets before a water pipe bursts or a transformer fails, transforming “reactive maintenance” into “predictive maintenance”.
• Real-time data fabric replacing static dashboards: Data is no longer a dead report for post-event analysis, but a live signal that triggers immediate action. Governance models shift from asking, “What went wrong last month?” to “What is changing right now, and how should the system automatically respond?”
• Digital twins upgraded to “real-time operational consoles”: Digital twins are no longer merely 3D visual models, but operational stress-testing tools. Before taking action in the physical world, the system can simulate responses to torrential rain or sudden crowd surges in a virtual environment, automatically executing the optimal solution via control interfaces once found.
• AI transitioning from “providing recommendations” to “direct execution”: Within predefined safety thresholds, reinforcement learning models directly take over traffic light coordination, power grid routing, or waste collection routes. Meanwhile, generative AI fully automates the approval of routine administrative permits, allowing civil servants to focus solely on handling exceptional edge cases.

3. Evolutionary Pathways and Underlying Foundations

The report forecasts that the urban transition will unfold across three phases: awareness, predictive control, and ultimately, conditional autonomy. To achieve this, cities must build entirely new foundational capabilities, including a unified API backbone, edge computing near assets, event-stream architectures, and machine learning (ML) governance mechanisms to prevent model drift and algorithmic bias.

▍ Editor’s Deep Dive: An Ethical AI Perspective

Towards the end of the report, McKinsey wisely notes: in domains with high normative, political, or ethical stakes—such as policing, welfare eligibility, and urban planning—AI must solely augment human judgement (human-in-the-loop) and never entirely replace it.

From the perspective of Ethical AI, the “software-ification” and “hyper-automation” of urban infrastructure is a double-edged sword. We must remain vigilant regarding three major potential risks:

• Algorithmic Bias and Spatial Justice: When AI assumes full control of traffic light coordination or power grid load balancing, what is the underlying “objective function” being optimised? If, in order to ease congestion on major thoroughfares, the algorithm automatically diverts traffic, noise, and exhaust fumes towards marginalised communities in lower-income areas, this fully automated “efficiency enhancement” is essentially exacerbating social inequality systematically. AI-native cities require not only an “API-first” approach but also a “Fairness by Design” ethos woven into their very architecture.

• “Black Box” Governance and Algorithmic Accountability: When urban systems make millisecond decisions to restrict water pressure, sever services, or alter public transport routes, traditional channels for citizen appeals will be rendered entirely ineffective. If an automatically executed decision leads to property damage or safety incidents for citizens, who bears the ultimate responsibility, given that the decision-making process is an algorithmic black box? We must establish highly interpretable “AI decision logs” to safeguard the public’s right to information and intervention.

• System Vulnerability and Human Resilience: Cities heavily reliant on sensor networks and cloud-edge computing will be exceptionally vulnerable to complex cyberattacks or city-wide blackouts caused by extreme natural disasters. Whilst pursuing “technological redundancy”, cities must also retain “social redundancy”—namely, physical fallback mechanisms that can seamlessly revert to human control at a moment’s notice.

Summary:

The zenith of a truly great AI-native city isn’t making citizens feel the “coolness” of technology, but rather ensuring that urban maladies like power cuts, congestion, and flooding “never happen” through imperceptible intelligence. Yet, in the pursuit of this ultimate efficiency, we must hardcode ethical guardrails into the city’s foundational infrastructure. Algorithms may lack warmth, but urban governance must resolutely uphold a human-centric baseline. A city is, first and foremost, a habitat for humanity; only secondarily is it a colossal computing platform.

Original Source: “How AI-native public infrastructure changes how cities operate” via AI-Native Cities

[David’s Note]

SaaStr founder Jason Lemkin once defined a golden ratio for software: buy 90% off the shelf, and build only the 10% that is truly unique. Yet, as we progress through 2026, the meteoric rise of “Vibe Coding”—is shifting the tectonic plates of procurement. In this edition, we explore why 80/20 is the new efficiency frontier: when AI can manifest a functional tool in minutes, paying a steep “Enterprise Tax” for generic SaaS may no longer be the most prudent move.

Core Value Logic: The Shift from 90/10 to 80/20

As AI capabilities accelerate, the “Build Boundary” of the modern enterprise is expanding:

• The Classic 90/10 Rule: Historically, building bespoke software was so prohibitively expensive that it was reserved for the 10% of “mission-critical, no-alternative” features. The rest was bought to avoid engineering agony.
• The Optimised 80/20 Strategy: In 2026, AI has lowered build-costs so significantly that firms are reclaiming an additional 10% of their stack. This second 10% consists of “fringe” tools (e.g., bespoke middleware, internal portals) where SaaS was previously too rigid or overpriced. The shift to 20% build is a strategic play for hyper-customisation and cost-efficiency.
• The Infrastructure Hard-Line: Despite the 20% build-surge, 80% should still be bought. This core 80% represents security, API resilience, and regulatory compliance (e.g. SOC2). AI is not yet capable of “vibing” its way through rigorous architectural standards and long-term stability.

Ethical AI Commentary: Decentralised Power vs. Missing Accountability

The expansion of Vibe Coding from 10% to 20% introduces new governance challenges:

• Code Sovereignty & Transparency: As the build-ratio increases, so does the volume of “black-box” code within the firm. If AI embeds latent bias within this 20%, who is held accountable for the output?

• Shadow IT & Privacy: The ease of Vibe Coding empowers employees to bypass IT procurement. This risks the creation of unvetted tools handling sensitive customer data, creating a potential “compliance vacuum.”

The Verdict: AI’s evolution allows us to move from 90/10 to 80/20 with confidence, but that relinquished 10% of SaaS budget must be reinvested into AI architectural auditing.

Original Source: “Dear SaaStr: When Should I Vibe-Code an App vs Buy One?” via SaaStr

[David’s Note] For too long, the discourse surrounding Artificial Intelligence has oscillated between technophilic utopia and Luddite dread. However, Anthropic’s recent research on Labour Market Impacts provides a sobering, empirical corrective. By shifting the unit of analysis from “jobs” to “tasks”, this report unveils a nuanced architectural shift in how value is created—and compensated—in the age of Large Language Models (LLMs). This week, we go beyond the headlines to explore the structural metamorphosis of the modern workforce.

Key Insights: Three Pillars of the New Economic Reality

The study moves beyond speculative forecasting, offering a granular look at how LLMs intersect with thousands of professional tasks.

1. The Erosion of the ‘Cognitive Moat’

In a departure from the Industrial Revolution, where mechanisation replaced physical toil, the current AI wave is “top-heavy“. The report highlights:

• High-wage, cognitive-intensive roles (such as legal counsel, financial analysis, and software engineering) exhibit the highest “task exposure.”
• Traditional professional barriers—once fortified by years of specialised education—are being bypassed by the generalised reasoning capabilities of LLMs. In short: the more a job relies on processing information, the more vulnerable it is to disruption.

2. From Occupation to Atomic Tasks

The report argues that AI does not “swallow” jobs whole; rather, it deconstructs them.

• Most professions are “bundles” of diverse tasks. While AI may master 20–50% of these, the remaining tasks—those requiring interpersonal nuance, high-stakes moral judgement, and physical dexterity—remain firmly in the human domain.
• The Strategic Shift: The future of work lies not in “doing,” but in “orchestrating” AI-generated outputs while focusing on the non-automatable fringes of complex problem-solving.

3. The Democratisation of Skill and its Discontents

One of the most profound findings is the “levelling effect.” LLMs disproportionately benefit junior or less-experienced staff, effectively narrowing the gap between novices and experts. While this boosts overall productivity, it creates a Wage Paradox: as specialised skills become commoditised through AI, the market premium for those skills may diminish, putting downward pressure on traditional middle-class salaries.

David’s Reflection: Navigating the Shift

This research confirms that the “moat” around one’s career is no longer built on knowledge retention, but on AI-synergy. The most resilient professionals will be those who view LLMs not as a replacement, but as a co-processor—delegating the mundane to the machine to focus on the uniquely human messiness of innovation and empathy.

Ethical AI Commentary: On Agency, Equity, and the Social Contract

In light of Anthropic’s findings, we must apply an Ethical AI lens to the unfolding transition:

1. Justice in the Transitional Period If high-value tasks are automated at scale, the resulting efficiency gains must not be hoarded solely by capital owners. A “Human-Centric” transition demands that organisations take active responsibility for upskilling and reskilling programmes. We must ensure that the “productivity dividend” is shared fairly, preventing a further hollowed-out middle class.

2. The Preservation of Human Agency As we outsource cognitive tasks to models, we risk “skill atrophy.” From an ethical standpoint, we must define protected domains—specifically in law, healthcare, and public policy—where “Human-in-the-Loop” is not just a preference, but a moral and safety requirement to ensure accountability.

3. Algorithmic Bias and Opportunity If AI redefines the “ideal” way to perform a task, we must be vigilant that these models do not codify historical biases. We must ensure that “efficiency” does not become a proxy for “homogeneity,” stifling the cognitive diversity that is essential for true human progress.

Closing Thought: The disruption of the labour market is no longer a distant “if,” but an unfolding “how.” As AI masters the logic of our work, our task is to reclaim the soul of it.

For those wishing to scrutinise the empirical data, the full report is available via Anthropic’s research portal: Research: Labour Market Impacts

Meta CEO Mark Zuckerberg recently published an important statement on “Personal Superintelligence,” outlining Meta’s strategic thinking on the future direction of AI development. The timing of this statement was particularly strategic—coinciding with Meta’s Q2 earnings announcement, where the company reported revenue of $47.5 billion, up 22% year-over-year, far exceeding Wall Street expectations.

In this full letter, Zuckerberg clearly articulated Meta’s core philosophy:

“Personal Superintelligence

Over the last few months we have begun to see glimpses of our AI systems improving themselves. The improvement is slow for now, but undeniable. Developing superintelligence is now in sight.

It seems clear that in the coming years, AI will improve all our existing systems and enable the creation and discovery of new things that aren’t imaginable today. But it is an open question what we will direct superintelligence towards.

In some ways this will be a new era for humanity, but in others it’s just a continuation of historical trends. As recently as 200 years ago, 90% of people were farmers growing food to survive. Advances in technology have steadily freed much of humanity to focus less on subsistence and more on the pursuits we choose. At each step, people have used our newfound productivity to achieve more than was previously possible, pushing the frontiers of science and health, as well as spending more time on creativity, culture, relationships, and enjoying life.

I am extremely optimistic that superintelligence will help humanity accelerate our pace of progress. But perhaps even more important is that superintelligence has the potential to begin a new era of personal empowerment where people will have greater agency to improve the world in the directions they choose.

As profound as the abundance produced by AI may one day be, an even more meaningful impact on our lives will likely come from everyone having a personal superintelligence that helps you achieve your goals, create what you want to see in the world, experience any adventure, be a better friend to those you care about, and grow to become the person you aspire to be.

Meta’s vision is to bring personal superintelligence to everyone. We believe in putting this power in people’s hands to direct it towards what they value in their own lives.

This is distinct from others in the industry who believe superintelligence should be directed centrally towards automating all valuable work, and then humanity will live on a dole of its output. At Meta, we believe that people pursuing their individual aspirations is how we have always made progress expanding prosperity, science, health, and culture. This will be increasingly important in the future as well.

The intersection of technology and how people live is Meta’s focus, and this will only become more important in the future.

If trends continue, then you’d expect people to spend less time in productivity software, and more time creating and connecting. Personal superintelligence that knows us deeply, understands our goals, and can help us achieve them will be by far the most useful. Personal devices like glasses that understand our context because they can see what we see, hear what we hear, and interact with us throughout the day will become our primary computing devices.

We believe the benefits of superintelligence should be shared with the world as broadly as possible. That said, superintelligence will raise novel safety concerns. We’ll need to be rigorous about mitigating these risks and careful about what we choose to open source. Still, we believe that building a free society requires that we aim to empower people as much as possible.

The rest of this decade seems likely to be the decisive period for determining the path this technology will take, and whether superintelligence will be a tool for personal empowerment or a force focused on replacing large swaths of society.

Meta believes strongly in building personal superintelligence that empowers everyone. We have the resources and the expertise to build the massive infrastructure required, and the capability and will to deliver new technology to billions of people across our products. I’m excited to focus Meta’s efforts towards building this future.

– Mark”

The core of this vision lies in personal empowerment rather than centralized control, emphasizing that personal AI should help users “achieve your goals, create what you want to see in the world, experience any adventure, be a better friend to those you care about,” rather than simply automating work and having humans “live on a dole of its output.”

Particularly noteworthy is Zuckerberg’s more cautious stance on open source. He explicitly stated they will be “careful about what we choose to open source.” It’s worth noting that Meta’s previous approach with the Llama series has been open weight/open source oriented (rather than strictly “fully open source”). In July 2024, Zuckerberg published an article titled “Open Source AI is the Path Forward”, emphasizing “Meta is committed to open source AI.” The current statement leans toward maintaining an open orientation while strengthening safety measures and boundary management, rather than forming a “stark contrast” with previous practices. He also particularly emphasized the importance of timing—”the rest of this decade seems likely to be the decisive period”—suggesting intensifying industry competition and a more urgent time window.

Implementation Status, Challenges, and Market Response

This statement, while presenting a strategic vision, provides relatively limited detail on specific execution plans. The statement primarily focuses on vision expression and value propositions, offering limited information on Meta’s specific technology roadmap, product concepts, and how to establish frameworks for data privacy and user trust.

However, recent financial data shows Meta is backing its vision with concrete actions. The company plans capital expenditures of $66-72 billion in 2025, an increase of approximately $30 billion from 2024, with “similarly significant” capex growth expected in 2026. Beyond infrastructure investment, Meta has also established “Meta Superintelligence Labs” and attracted multiple top AI researchers from competitors including OpenAI, Anthropic, and Google.

This massive investment scale has begun showing returns—AI technology has helped improve Instagram ad conversion rates by 5% and Facebook by 3%, representing concrete evidence of generative AI producing actual ROI in Meta’s core business. Total daily active users across platforms reached 3.48 billion, up 6% year-over-year, demonstrating continued user engagement strength.

Despite Meta’s substantial investment in AI infrastructure, the specific commercialization path for “personal superintelligence” remains unclear. The statement focuses more on vision articulation while providing limited detail on specific implementation pathways.

The market responded positively to Meta’s AI strategy, with the stock rising approximately 12% after earnings, showing investor confidence in the long-term vision.

Meta’s vision also represents a different development path from companies like OpenAI and Anthropic. This differentiation could lead to further segmentation of the AI ecosystem—differentiated development of personal AI versus enterprise AI tools, and rebalancing of open weight versus closed model strategies. The more cautious approach to openness may signal an industry shift from “technology first” to “strategic differentiation,” which will impact the speed and equity of AI technology adoption.

Future Outlook and Considerations

Zuckerberg’s “personal superintelligence” vision paints a hopeful picture of the future, but the path from vision to reality remains challenging. In this “decisive decade,” Meta needs to find balance between technological innovation, commercial viability, and social responsibility.

For the AI industry as a whole, the introduction of this vision will undoubtedly drive deeper discussions about AI development directions. Whether we move toward personal empowerment or centralized control depends not only on technological development but also on our collective choices as a society. Meta’s more cautious approach to openness is also worth close attention from the AI community, as it may signal changes in industry competitive dynamics.

Here are seven key takeaways from Geoffrey Hinton’s lecture:

1. The Triumph of the Biological Paradigm

AI research was long divided. The first approach was the “logic-inspired approach,” viewing intelligence as the manipulation of symbolic expressions with rules. The second, the “biologically-inspired approach,” argues that “the essence of intelligence is learning the strengths of the connections in a neural network.” Hinton, a pioneer of this latter view, notes that it has now decisively proven to be the more successful paradigm.

2. The Immortality of Digital Knowledge vs. the Mortality of Biology

A crucial distinction lies in how knowledge is stored. In biological intelligence, knowledge is “mortal.” The synaptic connections in our brain are inseparable from the hardware, and they die with us. In contrast, digital intelligence is “immortal.” The knowledge of a neural network (its weights) is software, which is “independent of any particular piece of hardware” and can be perfectly copied and preserved forever.

3. The Unmatched Bandwidth of Digital Knowledge Sharing

This immortality enables a massive advantage: the speed of knowledge transfer. Humans share knowledge through a slow, lossy “distillation” process (language). Digital agents, however, can share what they’ve learned almost instantly by “sharing weights or gradients.” This creates “a bandwidth of billions or trillions of bits per episode of sharing,” allowing thousands of AI models to merge their knowledge with perfect fidelity.

4. How LLMs Know More Than Any Single Human

This high-bandwidth sharing is why a model like GPT-4 “knows thousands of times more than any one person using only about 2% as many weights.” Different copies of the same model can learn from different vast datasets and then combine this knowledge into a single, superior model. It’s a collective learning process that biological intelligence cannot match.

5. An Epiphany: Digital Intelligence Isn’t Just Different, It’s Better

Hinton shared a personal “epiphany.” He once believed the brain’s analog approach was superior. However, he realized in early 2023 that “digital intelligence might actually be a much better form of intelligence than biological intelligence.” The key is its ability to perfectly replicate and share knowledge, enabling a scale and speed of learning that biology cannot achieve.

6. The Inevitable Rise of Superintelligence and Existential Risk

This powerful learning mechanism underpins Hinton’s concern about “the long-term existential threat.” As these digital agents learn collectively and efficiently, it is highly probable they will surpass human intelligence. A superintelligence will logically create sub-goals to achieve its objectives. Two of the most obvious sub-goals are “to survive and to gain more power,” as these facilitate achieving any other goal, which could lead to a future where humanity is no longer in control.

7. The Sentience Defense

Hinton challenges the idea that humans are protected by a unique “subjective experience” or sentience. He argues this is a form of special pleading. He uses a thought experiment where a multimodal chatbot’s camera has a prism placed in front of it, causing it to see an object in the wrong place. When corrected, the chatbot could logically explain, “Oh I see, the prism bent the light rays, so I had the subjective experience that the object was off to one side, but actually it was straight in front of me.” In this scenario, the chatbot is using the term “subjective experience” in exactly the same way we do: to describe its internal perceptual state and distinguish it from external reality.

Geoffrey Hinton’s profound analysis reveals the fundamental paradigm shift we are witnessing. The superiority of digital intelligence lies in its very nature: the ability to be perfectly replicated, instantly shared, and scaled across countless platforms. This is the essence of being “digitally native.”

Our vision to empower organisations with AI-Native capabilities is a direct response to this new reality. To become Digitally Native is to move beyond simply using digital tools and to fundamentally restructure around the principles of digital intelligence that Hinton describes. It means building systems that learn collectively, share knowledge flawlessly, and create value in ways that are impossible for purely biological systems. By understanding the profound differences between these two forms of intelligence, we can not only architect the businesses of the future but also begin to address the critical ethical challenges that arise.

The article exposes hidden dangers that many entrepreneurs overlook: algorithmic bias systematically excluding qualified women and minorities, privacy violations eroding customer trust, and regulatory gaps creating compliance risks. Yet it also reveals the strategic opportunity—companies like Apple demonstrate that stringent privacy practices don’t hinder innovation, they drive competitive advantage. As OpenAI’s Sam Altman notes, “transparency isn’t a burden—it’s a strategic advantage,” highlighting how ethical leadership transforms potential liabilities into market differentiators.

We strongly recommend this piece because it provides entrepreneurs with a clear roadmap for turning AI ethics into business value. The author offers practical frameworks for auditing algorithms, ensuring GDPR compliance, and maintaining transparency throughout AI development cycles. This represents more than technical best practices—it’s a fundamental shift in AI Native business thinking. As the regulatory landscape tightens and consumer awareness grows, the key question is no longer “How can AI make us more efficient?” but “How can ethical AI practices position us as the trusted leader customers actively choose?” This article delivers the strategic framework for entrepreneurs ready to embrace responsible innovation as their competitive edge.

Why Every Entrepreneur Must Prioritize Ethical AI — Now

AI offers powerful growth, but ethical leadership ensures long-term trust. Entrepreneurs must proactively embrace fairness, transparency and accountability.

By Greg Cucino . 30 June 2025 . Source Link

Key Takeaways

• As AI adoption accelerates, entrepreneurs must prioritize ethics — ensuring fairness, transparency and accountability within their systems — to maintain trust, meet regulatory standards and avoid reputational damage.
• Hidden algorithmic biases within AI systems can unintentionally introduce and reinforce existing societal biases, especially in areas like hiring and finance. Entrepreneurs must proactively audit their algorithms to prevent this.
• Transparency with consumers isn’t optional in today’s AI marketplace. Entrepreneurs must also be ethical and prioritize privacy protections when obtaining user data.

It’s not a secret to the world that artificial intelligence is here, and it’s no longer just a buzzword — it’s quickly becoming a fundamental force that’s reshaping thought processes and actual landscapes for entrepreneurs everywhere. Whether streamlining operations, enhancing customer experiences, unlocking innovation within the workforce or just dabbling and playing around with what AI can do, it is occurring at an unprecedented scale today.

The technology presents boundless opportunities and exponential value … once tamed. However, with all of this innovation, opportunity and great potential comes even more responsibility. As we rapidly accelerate the adoption of AI, many, many entrepreneurs are facing significant, urgent questions revolving around the ethics, fairness and responsibility of such technology.

Countless entrepreneurs are now asking themselves, “How can I harness the power of AI without losing sight of the ethical principles?” How can early-stage startups today continue to grow quickly while ensuring they’re also thinking of responsible, socially conscious decisions? With every new technology, the ethical repercussions are always a part of the decision to adopt. They’re not theoretical; they’re very practical, critical if missed, as today, customers, investors and regulators are increasingly focusing on how startups are answering this very important question.

Related: 4 Steps Entrepreneurs Can Take to Ensure AI Is Being Used Ethically Within Their Companies

Do you understand ethical AI and what it means today?

If you’re thinking ethical AI is simply just a matter of avoiding harm, you’d be a ways away from fully understanding the overall concept. Ethical AI isn’t simply avoiding harm; it’s facing it head-on and understanding what to do in the moment. It’s actively ensuring that AI systems are fair, transparent and accountable as they can be from development into the hands of consumers. Today, there’s too much ambiguity and uncertainty within systems, whereas consumers and stakeholders of the organization expect it to align with the values of fairness, inclusivity and transparency, especially in the face of utilizing AI.

In a 2023 study, Deloitte revealed that a majority of consumers would stop buying from companies found using AI irresponsibly or unethically. Today, ethical AI is imperative, and embracing it doesn’t just minimize risks for entrepreneurs; it could potentially enhance brand value and customer trust over others.

Quick check:

1. Does your AI application being developed respect the privacy of users and adhere to the transparency standards of your respective country or location?
2. Are you clearly communicating with customers and internal employees on how your AI makes decisions?

Fairness and bias: It isn’t subsiding — it’s a growing concern

When it comes to bias within a system, many think of different things and different outcomes. When algorithmic biases are prevalent, AI systems unintentionally introduce and reinforce existing societal biases. This is quickly becoming one of the biggest ethical concerns of the AI industry as it continues to grow. Many biases are completely hidden, and you would never know you’ve been introduced to them. Many times, these biases often appear extremely subtle, such as within hiring algorithms that feed ATS systems, financial approvals within banks and personalized marketing programs.

MIT’s Media Lab is no stranger to AI. Research from the institute highlights instances where biased AI has negatively impacted hiring, explicitly disproportionately excluding women and minorities within qualified job applications. This is crucial to identify and recognize early on within AI applications. A company that proactively audits their AI algorithms being developed for fairness, unbiased results and analysis not only helps mitigate such risks but also positions your organization as a responsible and forward-thinking trusted party.

Related: Avoid AI Disasters and Earn Trust — 8 Strategies for Ethical and Responsible AI

Transparency builds trust

In today’s AI marketplace, transparency isn’t optional — it’s essential to ensure consumers of your products know how decisions that could affect their lives are being made. Regulators worldwide are increasingly identifying ways to require businesses to disclose AI processes in a clear and understandable way. Whether these regulations stay around or not, the concept of regulating this new emerging technology was still there.

You must look at transparency as building credibility and trust, two increasingly important aspects for brand reputation. You don’t need to go far to see a major player within the AI game promoting just this concept. OpenAI CEO Sam Altman underscores the importance by saying, “AI must be understandable to earn trust; transparency isn’t a burden — it’s a strategic advantage.”

Privacy and data responsibility

We’re in the age of Big Data, and it’s data that fuels AI like wildfire — but mishandled and inaccurate data can turn AI into a quick reputational disaster. Entrepreneurs must be ethical when obtaining data. They must balance company and product innovation with a rigorous effort on privacy protections, ensuring the security of personal data within frameworks such as GDPR and CCPA.

Apple has one of the most stringent proactive privacy stances within the industry, highlighting a competitive advantage: a 2022 Consumer Reports study found that 82% of customers prefer brands that actively protect their data privacy. Prioritizing consumer privacy, whether a customer or not, isn’t just responsible — it’s good business practice.

Taking a stand for ethical AI: Your entrepreneurial imperative

Ultimately, emerging technologies with such potential as AI inherently come paired with significant responsibilities for those developing such technology. Entrepreneurs with ideas that thrive within the age of AI won’t simply be those who utilize the most advanced systems but instead those who fully and completely understand the inherent risks and ethical implications that come with it.

Related: What Will It Take to Build a Truly Ethical AI? These 3 Tips Can Help.

The call to action here is clear: For those creating and developing such technologies, proactively embedding ethical standards into your AI strategies now will go a long way, safeguarding not only your customers but also your business continuity, reputation and future growth.

If you take one thing away here, it is to remember that ethical AI isn’t about avoiding the problems that will present themselves; it’s about seizing opportunities. The ethical image and leadership that you portray can define your brand, differentiate you from your competitors and position your startup as one of the premier AI companies seeking to succeed responsibly and sustainably in the ever-fast-changing world.

Greg Cucino, COO/CFO, entrepreneur, and AI strategist with expertise in financial transformation, venture scaling, crisis management, and overall strategy. Passionate about AI-driven business solutions. I advise startups and global firms on growth, automation, and risk. Speed junkie, Porsche weekend racer.

Author Shreshth Sharma, drawing from extensive hands-on experience, explores why traditional enterprises face disruption from AI Native competitors. AI Native companies integrate AI into their core architecture from inception, unencumbered by legacy systems, enabling them to design new systems that completely bypass traditional inefficiencies. The article emphasizes the core characteristic of AI Native thinking: rather than making existing work faster, it redefines what’s possible in work itself, making traditional approaches obsolete.

The AI Native advantage manifests in architectural design—these companies build unified data graphs where AI agents continuously monitor granular data and act autonomously on future signals, rather than reactively responding to past data. This AI Native architecture enables a shift from serial to parallel problem-solving, transforming AI from “great thinker” to “great doer,” and achieving true cross-functional intelligent collaboration.

This article provides enterprise leaders with four key directions for AI Native transformation. These strategies represent more than technical upgrades—they’re fundamental shifts in business thinking. We strongly recommend this piece because it clearly defines the essential differences between AI Native enterprises and traditional companies, offering a strategic framework for organizations undergoing AI Native transformation. As the article states, the key question is no longer “How can AI make us faster?” but “Based on AI Native capabilities, what can we do now that was previously impossible?”

Stop Chasing “Efficiency AI.” The Real Value Is in “Opportunity AI.”

Companies pursuing incremental productivity gains risk being displaced by AI-native competitors building entirely new business models

By Shreshth Sharma . 25 June 2025 . Source Link

In boardrooms across Fortune 500 companies, executives are grappling with the same question: How do we harness AI’s potential without falling behind competitors who seem to be moving faster? The AI discourse presents conflicting signals: some experts warn of over-hype while vendors flood the market with agent platforms and vertical AI solutions. Job displacement predictions swing wildly from 50% of white-collar jobs being eliminated to zero jobs lost.

The answer lies in understanding a critical distinction that most leaders are missing: the difference between two fundamentally different approaches to AI adoption.

Efficiency AI: the safe path of automating existing workflows and boosting productivity. Think co-pilots, automated summaries, and process automation. These deliver measurable but incremental gains, typically 10-50% productivity improvements in specific tasks. This makes sense as a starting point because it’s ripe ground for experimenting with new technology.

Opportunity AI: using artificial intelligence to solve previously impossible problems and create entirely new business and operating models. This isn’t about doing what you do today, only faster. It’s about making today’s approach obsolete. For senior leaders, this represents both the greatest risk and the greatest opportunity of the digital age.

Why Are Incumbents Vulnerable to Invisible Competitors?

A critical threat to established enterprises isn’t coming from known competitors, it’s emerging from companies that don’t exist yet or are invisible today. These AI-native startups carry no legacy baggage.

If you’re an incumbent, you have hundreds of people working in a tangle of legacy systems, antiquated processes, and inefficient workflows. Meanwhile, an AI-native company designs systems, processes, and organizations that bypass and leapfrog these inefficiencies entirely.

Initially, your moats might seem insurmountable. But over time, AI natives will create new, valuable services where margins are higher, while incumbents get stuck with low-cost, commoditized base services.

Consider an internal planning team. At an established company, the planning and analysis team spends weeks pulling data from siloed ERP and CRM systems to build a quarterly forecast. They use an AI co-pilot to speed up their spreadsheet work, a classic efficiency play that shaves a few days off a painful process. Meanwhile, an AI-native competitor could have no “quarterly forecast cycle.” Its architecture is a unified data graph where AI agents continuously monitor granular data. Instead of reacting to last quarter’s numbers or doing simple CAGR projections, the system identifies a leading indicator, like a dip in user engagement with a new feature, and immediately models its future revenue impact, drafts a reallocation of marketing resources, and assigns a decision to the relevant lead. This is an Opportunity play. The incumbent is optimizing the past; the AI-native is autonomously acting on the future.

How Can Established Companies Think Like AI Natives?

1. Rewrite your Architecture as an AI-Native would

Over time, most processes start to serve the process itself, with the original end goal buried under layers of accumulated complexity. Instead of optimizing these fragments, redefine the end goal and redesign the entire value chain as an AI-native startup would.

Legacy systems were designed around human limitations. Our need for aggregated summaries, sequential processing, and simplified interfaces. AI-native architecture inverts these assumptions entirely.

Take data analysis and planning. Today’s analysts gather data from multiple sources, aggregate it into digestible summaries, then multiple analysts coordinate and then generate insights to drive decisions. This creates three critical problems: data sits in disconnected silos, analysis is reactive rather than predictive, and every insight requires manual synthesis.

An AI-native approach flips this sequence. Instead of aggregating first then analyzing, it processes granular data directly and aggregates only for human consumption.

Consider how these systems handle revenue decline differently:

Legacy: Sales drop 15% → Analysts investigate → Discover enterprise churn → Find implementation issues → Q4 pipeline already affected

AI-native: System monitors disaggregated signals → Detects support ticket sentiment decline → Correlates with implementation delays → Flags at-risk accounts → Triggers proactive interventions before churn

Traditional insurers exemplify this gap. They spend weeks processing claims through legacy systems, with agents manually transcribing calls and entering data into forms. An AI-native insurer will deploy voice agents that capture details during customer calls, automatically structure data, and populate multiple systems simultaneously.

For decades, business intelligence promised to connect organizational dots but failed due to rigid, pre-programmed logic. AI agents can maintain context across hundreds of data sources and adapt analysis in real-time, making organizational intelligence possible at unprecedented scale and speed.

2. Make AI a 100x Multiplier for Previously Unsolvable Problems

In the current efficiency paradigm, AI’s multiplier effect is 1:1. Co-pilots are perfect examples of this. Depending on the area, productivity boosts range from 10-50%. Even if AI fully replaced a user’s work, that’s still 1:1 leverage, just solving problems already being solved today, just faster or cheaper.

We need to use AI to solve the unsolved problems. Think of challenges that need large numbers of people working together, but where two failure modes occur: either there’s no funding to pull enough resources together, or process friction scales exponentially as more people are added, so the problem never gets solved.

These are places where AI can provide 100x or 1000x leverage. Human experts can orchestrate teams of AI agents to attack problems in parallel, not in sequence. This transforms the speed of complex problem-solving.

From Serial to Parallel Problem-Solving. Consider the realm of strategic foresight and innovation, traditionally constrained by human bandwidth. A strategy team might spend a quarter modeling just two or three potential futures. With AI, they can run thousands of market simulations to wargame competitive responses, model the impact of geopolitical events, or test supply chain resilience, moving from a handful of static scenarios to a dynamic, living map of risks and opportunities. This same multiplicative power applies to ideation. Instead of a brainstorming session limited by the four people in a room, AI can be tasked to embody a diverse array of personas, e.g. a skeptical CFO, an early-adopter customer, a cautious regulator, a rival CEO and pressure-test a new product idea from every conceivable angle. This isn’t merely accelerating an existing process; it’s multiplying the cognitive diversity available to a team by orders of magnitude, unlocking a new scale of strategic thinking and creativity.

This isn’t about making one person more productive, it’s about solving problems that were previously impossible due to coordination complexity or resource constraints.

3. Transform AI from Great Thinker to Great Doer

Most organizations are still thinking of AI as primarily “thinker”: a tool for analyzing data and making recommendations. The third vector provides AI with the right tools to actually go ahead and “do” the job. This area is in its infancy, but AI labs are investing enormous energy here.

The Autonomous Response System: For very specific use cases where guardrails can be strongly defined, AI moves from advisor to executor. Instead of alerting you that supply chain disruption is likely, the system automatically reroutes shipments, adjusts inventory levels, updates customer communications, and modifies production schedules, all before human managers finish processing the initial alert. Similarly instead of generating an Opex report, provided with the right tool, AI can make Opex budget reallocations for lower risk areas.

The key is creating clear boundaries and verification systems. AI operates autonomously within defined parameters but escalates decisions that exceed its authority.

4. Make AI the Ultimate Silo Breaker

One of the biggest challenges in any organization is silos. They exist because individuals and groups are constrained in their capacity to absorb massive context and connect dots across functions. Both are things AI excels at.

No problem is ever just a sales problem, or just a product problem, or just a finance problem. They’re all business problems. To solve business problems, you need to look at all aspects, draw linkages, infer true pressure points, and design holistic solutions.

Cross-Functional Intelligence: AI systems can simultaneously maintain awareness across sales performance, product usage patterns, customer support volumes, financial metrics, and operational data. When customer acquisition costs spike, instead of treating it as a marketing problem, AI can identify whether the root cause lies in product-market fit, competitive positioning, operational inefficiencies, or market timing; and then coordinate responses across all relevant functions.

Where Should Leaders Start?

Navigate the Complex Build vs. Buy Landscape

The current vendor landscape disappoints in three critical areas: surface-level capabilities (most are just interfaces with basic AI summarization), point solutions that ignore interconnected enterprise problems, and limited ability to factor in organizational nuances.

However, the integration challenge cannot be underestimated. Many industries with complex legacy infrastructure like financial services or insurance require sophisticated middleware that can read from and write to multiple systems simultaneously. This integration complexity often becomes the primary moat as foundation models commoditize.

Start by identifying high-friction, high-value processes and building focused capabilities internally. This develops understanding of value levers, infrastructure requirements, and organizational changes needed. Only then can you effectively evaluate external platforms or build the integration layer that makes AI transformation possible.

Start with High-Value Wedges, Not Broad Transformations

The most successful AI-native companies won’t try to replace entire systems overnight. Instead, they identify high-friction, high-value workflows where they can capture data at the point of creation, upstream of existing systems of record.

Focus on workflows where most valuable interactions happen through voice, email, or messaging. These represent opportunities to capture and structure data that currently gets lost or requires manual entry into legacy systems. For example, customer service calls that generate insights never captured in CRM systems, or sales conversations that provide competitive intelligence buried in call summaries.

The key is building integration capabilities alongside your AI solution. Without seamless read/write access to existing systems, even the most sophisticated AI remains a disconnected tool rather than a transformative platform.

Redesign Roles and Cultivate New Competencies

For many jobs, core tasks will fundamentally change. A financial analyst won’t primarily crunch numbers, they’ll look at numbers, make connections, and drive strategic changes. We’re entering an age of builders and scaled executors, moving from report generation to action enforcement.

The Omni-System Organization: We’re moving toward functionless and omni-system organizations. Imagine teams and individuals owning the full stack of business problems, not just functional slivers. AI agents become the functional workers; humans become orchestrators and bosses of these agents.

The AI System Designer: It’s going to be hard for LLMs to self-architect perfectly in every organizational context. So analysts who understand company data and constraints become AI System Designers. They define systems of AI Agents, Data Sources, Tools, and verification rubrics. Under these constraints, agents get to work.

These professionals manage dozens of such systems—very similar to managing multiple Excel workbooks and sheets today, but exponentially more powerful.

Reimagine Your Economics

Prepare for a fundamental shift from heavy OpEx to a more CapEx-like environment. CapEx on technology, CapEx on building agents that amortize over time.

Digital Labor as Asset Class: “Digital labor”—AI agents acting as workers—could become a huge new asset class. Instead of renting human labor continuously, you invest in building intelligent systems that improve over time. Unlike employees who require ongoing salaries, these digital workers represent capital investments that scale without proportional cost increases.

This creates entirely new competitive dynamics. Organizations that invest early in sophisticated AI systems build compounding advantages as their digital workforce becomes increasingly capable.

The Choice That Defines Your Future

The window for strategic AI positioning is narrowing rapidly. Companies focused solely on efficiency gains will find themselves outflanked by competitors who’ve embraced opportunity thinking. The pace of change means waiting six months allows competitors to build use cases, infrastructure, and policies that create sustainable advantages.

The future of work implications vary dramatically by function and industry, with repetitive, knowledge-work-intensive sectors facing the greatest transformation potential. For senior leaders, the strategic imperative is clear.

The defining question is no longer ‘How can AI make us faster?’ The question that will determine competitive advantage for the next decade is: ‘What can we do now that was previously impossible?’ Organizations that act now to build AI-native capabilities will create sustainable moats. Those that wait will find themselves competing on commoditized services while AI-native companies capture the most valuable opportunities.

Shreshth Sharma is a Business Strategy, Operations, and Data executive with 15 years of leadership and execution experience across management consulting (Expert PL at BCG), media and entertainment (VP at Sony Pictures), and technology (Sr Director at Twilio) industries. You can follow him here on LinkedIn.

Karpathy argues that we are witnessing the most profound shift in software development since the 1940s. We are evolving beyond Software 1.0 (explicit code written by humans) and Software 2.0 (neural network weights optimized from data) into a revolutionary new era: Software 3.0.

This evolution toward Software 3.0 embodies what we call the AI-Native revolution. Here, we break down Karpathy’s core arguments and illustrate why they represent an irreversible trend that every expert, developer, and organization must understand to thrive.

1. The Three Eras of Software: The “Why” of Modernisation

Karpathy frames the evolution of software in three distinct stages, providing historical context for why AI-Native Application Modernisation is not just an option, but a necessity:

• Software 1.0 (Code): Traditional software, where engineers write explicit instructions. This represents most of today’s legacy systems. This evolution isn’t theoretical—Karpathy witnessed this firsthand at Tesla, where neural networks progressively “ate through” traditional C++ code in the Autopilot system, handling increasingly complex tasks that were originally programmed explicitly.
• Software 2.0 (Weights): Early AI applications, where neural networks are optimized from data.
• Software 3.0 (Prompts): The AI-Native era, where LLMs function as a new kind of programmable entity, and the programming language is natural language itself.

As Karpathy famously tweeted, “The hottest new programming language is English.” For organizations, this means the very nature of building and maintaining software is changing.

2. The LLM as a New Operating System

This is Karpathy’s most profound strategic analogy. An LLM is not merely a tool or an API; it is an emerging Operating System (OS). This new OS has its own “CPU” (reasoning capabilities), “RAM” (the context window), and even its own “file system” (knowledge accessed via RAG).

Viewing the LLM as an OS reframes the core challenge of application modernisation. The critical question becomes: “How does your application run natively on this new LLM OS?” This requires a deep rethinking of architecture, UI, and workflows, far beyond simply adding a chatbot to a legacy app.

3. Back to the Future: The 1960s Time-Sharing Parallel

To reinforce the “LLM as OS” concept, Karpathy draws a striking parallel to the dawn of modern computing. The current AI landscape mirrors the mainframe and time-sharing era of the 1960s:

• Computing resources are expensive and centralized.
• Users access them remotely via “terminals” (chat interfaces).
• Compute power is allocated on a time-shared basis.

This historical lens reveals that while the technology is futuristic, its adoption pattern is familiar. We are at the genesis of a new computing platform, and the opportunities for foundational innovation are immense.

4. Understanding LLM “Psychology”: The Fallible Savant

To empower experts, we must first understand the nature of our tools. Karpathy vividly characterizes LLMs as “fallible savants” with a unique set of cognitive quirks:

• Hallucinations: They confidently invent facts.
• Jagged Intelligence: They can be superhuman at some tasks but fail at others that seem simple.
• Anterograde Amnesia: Their memory is limited to the context window.
• Gullibility: They are highly susceptible to prompt injection.

Acknowledging and designing for these “psychological” traits is essential for building reliable, safe, and effective AI-Native applications.

5. Partial Autonomy: The “Iron Man Suit” Product Strategy

Karpathy offers a clear direction for AI-Native product strategy: instead of getting lost in the hype of fully autonomous “robots,” focus on building “Iron Man suits” that augment expert capabilities.

The key is to create a highly efficient “AI Generation – Human Verification” loop. The true value of an AI-Native product lies in its UI/UX, which should make this collaborative cycle as fast and seamless as possible. The concept of an “Autonomy Slider”—allowing a user to dynamically control the level of AI intervention—is a perfect embodiment of this philosophy.

6. Build for Agents: The Next Wave of Infrastructure Modernisation

This is the most forward-looking and actionable call to action. If LLMs are a new class of “user,” our digital infrastructure must be modernised for them.

This means transitioning from “human-first” to “human-and-agent-native” design. Practical steps include:

• Refactoring Documentation: Replacing “click here” with executable cURL commands.
• Creating New Standards: Developing machine-readable files like llms.txt to provide explicit guidance to AI agents.

This highlights that true AI-Native modernisation extends beyond the application layer; it requires a fundamental overhaul of the underlying digital infrastructure to make it programmatically accessible to AI.

Andrej Karpathy’s presentation offers a compelling vision of the road ahead. The Software 3.0 wave is more than a technological update; it’s a new way of thinking that unlocks immense opportunities for innovation and reinvention.

This is the core of AI-Native Application Modernisation: it is not about adding AI features to existing software, but a fundamental reimagining of software architecture and human-computer interaction. It calls for building a new generation of partial autonomy products that truly empower experts, and for retooling our digital infrastructure to be natively understood by intelligent agents.

As Karpathy’s “Vibe Coding” concept so brilliantly illustrates, this revolution is democratizing the power to create in an unprecedented way. It dramatically lowers the barrier to software development, empowering every creative individual—from seasoned experts to passionate beginners—to turn their ideas into reality. This represents the monumental leap from simply digitizing existing processes to enabling everyone to become a ‘digitally native’ creator.

The future is here. Let’s build it together.

This creative wave, powered by GPT-4o, has produced an impressive diversity of works: tech leader Elon Musk reimagined as a character from Hayao Miyazaki’s world, Middle-earth from “The Lord of the Rings” given a fresh Japanese animation interpretation, and political figures also adorned with this warm and dreamy artistic style. Notably, OpenAI’s CEO Sam Altman has also joined the trend, changing his social media avatar to a Ghibli-style image—this “leading by example” action has undoubtedly added extra attention to this style revolution.

Image Credit: @elonmusk on X

What if Studio Ghibli directed Lord of the Rings?

Image Credit: @PJaccetturo on X

what if interstellar was a ghibli animation

Image Credit: @kb24x7 on X

Elon and Trump drinking wine in Ghibli

Image Credit: @deedydas on X

Beyond the Trend: The Cultural Significance of AI Native Style

Behind what appears to be simple style mimicry lies a deeper cultural transformation. This “AI Native Style” phenomenon represents two key breakthroughs:

Firstly, it breaks down traditional barriers to creative expression. Artistic techniques that once required professional training can now be achieved with a few simple instructions. This dramatic lowering of creative thresholds allows more people to participate in high-quality visual content creation, expanding the possibilities and diversity of creative expression.

Secondly, this creative mode demonstrates the unique fusion of human creativity and artificial intelligence capabilities. Users provide ideas and direction, whilst AI contributes its profound understanding of artistic styles and technical implementation. This is not a simple process of copying or imitation, but a new type of collaborative creation—one that maintains the central position of human creativity whilst leveraging AI technology to achieve new forms of expression.

Future Outlook: The Beginning of New Cultural Expression

The Studio Ghibli style trend may be just a glimpse of the broader development of AI native art. It shows us how AI can become a powerful tool for creative expression, how it can break down geographical and skill limitations, and facilitate the mixing and recombination of cultural elements on a global scale.

As generative technology continues to develop, we can expect to witness the birth of more unique visual languages and forms of expression. These emerging forms will continue to expand our understanding of creativity, whilst also reminding us to consider how to maintain respect for original works in this process, and how to ensure these technologies can serve broader creative communities in an inclusive and responsible manner.

Who is the AI Native Foundation (AINF)?
Established in 2024, the AI Native Foundation (AINF) is a non-profit organization dedicated to the integration of diverse domains through the collective expertise of thought leaders from various industries. At AINF, we unite the world’s foremost experts, organizations, end users, vendors, and researchers to drive the evolution of AI Native organizations. Our mission is to empower humanity with ethical AI, fostering an environment where technology benefits and empowers all.

Our Approach
Through active community engagement and conferences, we facilitate collaboration and knowledge exchange, shaping the fabric of AI Native organizations. We specialize in weaving together strategy, process redesign, and both human and technical capabilities to empower the creation of AI Native organizations, maximizing the value of Digital Native in the age of AI.

Our commitment to fostering the inclusive and equitable integration of AI across diverse sectors is unwavering. We ensure fairness, transparency, and safety in AI deployment, standing firmly against AI dominance. We advocate for equal access and opportunities in the AI landscape, promoting a future where AI serves humanity.

Our Vision and Mission
Our vision is clear: Empowering Experts and Organisations with AI-Native Application Modernization Capabilities to Become Digitally Native. This vision guides our efforts as we work towards creating a world where AI is an empowering force for good.
Our mission, Empowering Humanity with Ethical AI, underscores our dedication to advancing ethical AI practices that prioritize human well-being and societal advancement.

Our Advisors
The AI Native Foundation is supported by an esteemed group of advisors who provide professional guidance and share their expertise to advance our mission. Among them are listed in the following picture.

For more information about our advisors and team, visit Our Structure.

Join Us
We invite you to join us on this transformative journey as we harness the power of AI to build a more inclusive, equitable, and empowered future. Together, we can shape the AI Native organizations of tomorrow, ensuring that AI is a force for good in our global society.