Category: Artificial Intelligence

Weekend Notebook #51 – when AI stops advising and starts acting

amitabh apte

Published on LinkedIn and amitabhapte.com on21st Dec 2025

This week in AI – from Intelligence to Agency

Something subtle but decisive shifted this week.

Not in capability.
Not in valuation.
But in intent.

Across security, payments, software creation, and capital allocation, AI is no longer being positioned as a decision-support layer. It’s being designed as an execution layer.

That distinction changes everything.

For years, AI sat comfortably in the advisory role. It analysed. It recommended. It optimised. Humans still pulled the final lever. That boundary is now eroding, not through a single breakthrough, but through quiet, cumulative design choices.

Consider cybersecurity. Google Cloud’s expanded partnership with Palo Alto Networks is not just a large services deal. It reflects a deeper truth. In an AI-shaped threat landscape, human-in-the-loop defence is too slow. Security systems are being rebuilt to detect, decide, and respond autonomously. Defence is becoming algorithmic by necessity, not ambition.

The same shift is visible in commerce. Visa’s AI agents completing real consumer purchases may sound incremental, but it marks a psychological crossing. When AI systems transact on our behalf, trust is no longer abstract. It’s operational. The question stops being “is this recommendation accurate?” and becomes “am I willing to let this system act for me?”

That leap from suggestion to execution is irreversible.

Capital markets are aligning to the same logic. SoftBank’s scramble to close the final tranche of its OpenAI commitment is not about hype or fear of missing out. It’s about securing influence over platforms that will increasingly do, not merely advise. Lightspeed’s $9 billion fund, alongside similar mega-raises, reflects the same recalibration. AI companies are no longer lightweight software plays. They are infrastructure operators, with execution risk, physical constraints, and balance sheets to match.

Even software creation itself is being reframed. Nvidia and Alphabet backing Lovable is not just about no-code tools. It’s about collapsing intent into output. When natural language becomes a production interface, creation shifts from specialised craft to conversational control. The bottleneck moves from technical skill to clarity of thought.

Across all of this, one pattern holds.

AI is migrating from intelligence to agency.
From knowing to doing.
From tools we consult to systems we delegate to.

And that is not a technical evolution. It is a leadership one.

My takeaway this weekend

We are crossing the line from assisted intelligence to delegated authority.

That transition is happening faster than most organisations realise, and far faster than governance, culture, or leadership muscle memory can absorb. AI systems are beginning to act inside workflows, markets, and creative pipelines that were designed for human accountability.

The risk is not runaway intelligence.
It’s misplaced trust.

The leaders who will navigate this next phase successfully won’t be the ones chasing the most autonomy, but the ones redesigning decision rights, escalation paths, and responsibility models for a world where machines execute at machine speed.

AI leadership is no longer about adoption.

It’s about delegation, and knowing exactly where not to delegate.

Beyond AI: my mindshare – The World Meditation Day

Today, on World Meditation Day, the contrast couldn’t be sharper.

We are building systems that move faster, decide quicker, and act without hesitation. At the same time, the human nervous system is struggling to keep pace. Cognitive overload is becoming the hidden tax of digital acceleration.

Meditation feels almost countercultural in this context, but that’s precisely the point.

With over 700 studies linking meditation to focus, emotional regulation, and resilience, it’s no longer a spiritual indulgence. It’s a leadership capability. In environments where decisions are increasingly automated, clarity of intent becomes the scarcest resource.

Stillness trains that clarity.

As AI systems take on more execution, the human role shifts upward, toward judgement, ethics, and meaning. Those are not skills we can rush or automate. They require space. Attention. Presence.

This isn’t about slowing progress.
It’s about strengthening the human core that guides it.

In an age of delegation, inner discipline becomes the final guardrail.

Weekend Notebook #50 – When Guardrails Drop and Hardware Stalls

amitabh apte

Published on LinkedIn and amitabhapte.com on14thDec 2025

This week in AI – The Great Contradiction

This week, the AI story fractured. Not because progress slowed, but because it accelerated unevenly.

Policy surged ahead. Models leapt forward. Infrastructure hit resistance.

What emerged was a stark contradiction at the heart of the AI economy: intelligence is scaling at digital speed, but deployment is still bound by physical reality.

Three signals made that tension unmistakable.

First, the velocity.

The US administration signalled a decisive shift. Speed now trumps caution. President Trump’s Executive Order blocked states from regulating AI, creating a federal fast lane for Silicon Valley.

The intent was clear: remove friction, accelerate advantage.

Markets responded immediately. OpenAI released GPT-5.2, not just a smarter model, but a professional-grade, agentic system designed for autonomy rather than conversation. This is AI built to act, not assist. The guardrails are thinning, and the models are accelerating.

This wasn’t coincidence. It was causality.

Second, the stall.

While software sprinted, infrastructure stumbled. Oracle shares dropped 11 percent on deployment delays, pulling Nvidia, CoreWeave, and Micron down with them. The reaction wasn’t about earnings. It was about execution.

The reminder was blunt: the Capacity Race is harder than the Capability Race. You can ship code overnight. You cannot pour concrete, secure power, or stabilise grids at the same pace. Physics still sets the tempo.

For leaders, this matters. AI advantage is no longer constrained by algorithms. It is constrained by land, energy, and logistics.

Third, the shift.

Disney invested $1 billion in OpenAI to license its characters for Sora. While others litigate,

Disney is operationalising. By moving its IP into generative video workflows, it validated Sora as a production-grade creative engine.

This isn’t just a media story. It’s a strategic pattern. IP owners are moving from defence to deployment, from protecting archives to activating them. The future of content is not about preservation. It’s about animation at scale.

My takeaway this weekend

We are watching infinite digital ambition collide with finite physical reality.

Policy is pushing. Governments are clearing the regulatory path.
Models are pushing. GPT-5.2 is ready for autonomous work.
Physics is pushing back. Infrastructure is now the bottleneck.

The constraint has shifted.

“The bottleneck is no longer policy or software. It is concrete and power. The winner in 2026 will not simply be the company with the smartest model, but the one that can physically deploy intelligence faster than everyone else. AI leadership is becoming an execution discipline.”

Beyond AI: my mindshare – The Faces Behind the Machine

I paused this week on the cover of TIME magazine. The “Person of the Year” wasn’t a single individual, nor was it AI itself, as the “Computer” once was in 1982. It was the Architects of AI: Altman, Huang, Zuckerberg.

That choice matters.

For years, we’ve spoken about AI as if it were weather. Something inevitable. Something happening to us. By putting human faces on the cover, TIME reminded us of a grounding truth:

AI is not weather. It is architecture.

It is the result of choices. Trade-offs. Incentives. Ego. Ambition.

Seeing these builders grouped together, competitors and collaborators at once, reinforced something easy to forget amid the abstractions of silicon and scale. The most powerful operating system shaping AI’s future is still the oldest one we have.

Human nature.

“As we head into the holidays, that’s both comforting and unsettling. The machines are learning fast. But the direction they take still depends on the people building them. And that responsibility hasn’t been automated away.”

Weekend Notebook #49 – When Media Learns to Live with AI

amitabh apte

Published on LinkedIn and amitabhapte.com on7thDec 2025

This week in AI – Deals, lawsuits, and the new media survival playbook

This week, traditional media finally showed its hand in the AI era.

Some chose partnership. Others chose confrontation. And together they revealed a simple truth.

Media is no longer negotiating with platforms. It’s negotiating with algorithms.

Meta struck multiple AI licensing deals with publishers, a pragmatic shift from conflict to compensation. Better to pay for data than fight over it.

Meanwhile, The New York Times went the other way, suing Perplexity for allegedly copying millions of articles. This isn’t just a copyright case. It’s a line in the sand.

“If AI becomes the new front page, who gets compensated for the journalism that trains it?”

Hollywood signalled its own adaptation curve. The $72B Warner Bros–Netflix pact is less about creative ambition and more about survival in an AI-enhanced production economy, where scale, tooling, and efficiency win.

Regulators made their presence felt too. Elon Musk’s X was fined €140M by the EU.

Australia signalled tougher accountability measures as social platforms edge closer to bans.

And then came a preview of the next frontier, Tilly Norwood, a fully AI-generated actress, no agent, no schedule, no limitations.

Cinema isn’t just being disrupted. Its economics are.

My takeaway this weekend

This was the week media stopped asking “What is AI doing to us?”
And started showing “How we plan to survive it.”

Some will partner.
Some will litigate.
Some will scale.
Some will reinvent.

But the pattern is unmistakable.

“AI is now the terrain media must learn to operate on. The winners will defend their value, license smartly, and adapt faster than algorithms evolve.”

Beyond AI: my mindshare – the FIFA Football World Cup Draw

Football’s biggest stage returns in 2026, stretching across the USA, Canada, and Mexico.

The World Cup isn’t just a tournament. It’s a global mood shift.

From Boston to Miami, Dallas to New York, iconic cities will host matches that celebrate talent, teamwork, and the world’s most universal language. England and Scotland will add their own chapters to the story, with fixtures that promise drama and emotion.

June to July will bring shared living rooms, late-night broadcasts, unforgettable goals, and a month-long reminder of why sport binds us.

The final at MetLife Stadium won’t just close a tournament. It will close a collective experience.

Weekend Notebook #48 – When AI gains momentum across global geographies

amitabh apte

Published on LinkedIn and amitabhapte.com on 30th Nov 2025

This week in AI – Growth beyond the West Coast, global players join the AI race

Some weeks, the AI narrative feels local, shaped by a handful of companies on the US West Coast.

This was Not one of those weeks.

This was the week AI revealed its global ambitions.
Not through model releases, but through cranes, concrete, capital, and policy.
Every region moved in its own direction, and for the first time, those directions felt equally significant.

Asia moved with industrial intent.

India’s Adani Group is reportedly in talks with Google on a multi-billion-dollar data-centre partnership. India is no longer content being the world’s digital back office. It wants to be a compute power in its own right.

Japan, meanwhile, approved Micron’s $96B semiconductor project, not a press release, but a generational bet on the Indo-Pacific becoming the world’s memory and materials corridor.

China took a different route.

Alibaba didn’t just talk about AI; it shipped it, Quark AI glasses and a cloud business now lifted by real workloads. China may not be leading the frontier-model race, but it is winning the deployment race. Intelligence at the edge, not just the cloud.

Africa showed the geopolitics beneath the ambition.
A WSJ investigation revealed Chinese firms now control nearly half of Africa’s IPv4 addresses. This is not an obscure technical footnote. In an AI-driven world, whoever controls the pipes controls the possibilities.

Europe leaned into its slow. steady. deliberate. advantage.

A new CNBC analysis suggests Europe’s measured approach may age better than Silicon Valley’s “move fast” doctrine. Markets seem to agree, Germany’s stock market is climbing again. Siemens Energy is emerging as one of Europe’s biggest beneficiaries of the AI-driven infrastructure boom with nearly a 100B Euro valuation.

The UK added its own twist: Qatar invested hundreds of millions into a British quantum startup, strengthening Britain’s emerging position in the quantum race.

And Revolut’s new $75B valuation underscores something deeper: global finance is quietly becoming one of AI’s most powerful accelerators, built on software, data, and automated risk intelligence.

The Middle East kept doing what it always does. Skip the pilot. Go straight to scale.

Uber’s driverless robotaxis will now operate commercially in the UAE. Not a test. A rollout. AI is becoming lived infrastructure faster than many Western cities can regulate it.

Zoom out, and one conclusion becomes impossible to ignore.

“AI is no longer a Western invention. It is a global construction project.”

My takeaway this weekend

AI is no longer a single race.
It’s a set of parallel movements shaped by regional ambition.

India and Japan are building the physical backbone. China is pushing intelligence into daily life. Europe is turning trust into strategy. The Middle East is deploying at city scale. Africa is emerging as the new arena for digital sovereignty. And the UK is betting on quantum, the foundational layer beneath it all.

The real question has shifted. Not “Who has the smartest model?”, but “Who can build the steadiest system around it?”

Compute, energy, policy, capital, trust this is where advantage now compounds.

AI leadership is moving from algorithms to infrastructure.
From speed to readiness.
From capability to resilience.

“The next winners won’t be the fastest movers. They’ll be the regions and leaders who can scale AI meaningfully across diverse geographies and real consumer impact.”

Beyond AI: my mindshare – the comfort of Christmas Films

This post comes just a day before December. Is it too early to think about Christmas? Perhaps. But there’s something timeless about the way certain films weave themselves into our holiday rituals.

This week, The Times published a beautiful guide to the best Christmas movies to stream, a reminder that stories can be as comforting as a warm blanket on a cold evening.

From the mischief of Elf and the tangled romances of Love Actually to the quiet grace of It’s a Wonderful Life, these films aren’t just entertainment. They are memory-makers. They carry the scent of mulled wine, the glow of fairy lights, and the soft permission to slow down.

What struck me most is how these classics endure despite infinite choice.

In a world of endless scrolling, we still return to the familiar to Bedford Falls, to Kevin McCallister’s misadventures, to the belief that hope can arrive in the most unexpected ways.“Perhaps that’s the real gift of the season: the chance to pause, revisit old stories, and let them remind us of what matte

Weekend Notebook #47 – When models mature and the world stretches

amitabh apte

Published on LinkedIn and amitabhapte.com on23rd Nov, 2025

This week in AI – Gemini 3 and the new infrastructure race

This was the week Google forced the AI narrative to tilt again. Not through hype, but through the release of Gemini 3, a model that signals a deeper shift in where the frontier now sits. For the first time in a while, the conversation wasn’t about clever demos or novelty features. It was about capability that feels embedded, a model designed to sit inside Google’s full ecosystem of search, cloud, devices, and productivity tools.

Gemini 3 lands as an integrated intelligence layer, not a standalone chatbot. And that matters. In AI, distribution consistently beats brilliance. Google’s advantage is not just the model. It’s the hundreds of millions of moments, queries, clicks, sessions, and decisions where that model can quietly shape outcomes.

But the more revealing signal came from inside Google itself: the acknowledgement that its AI-serving infrastructure must double every six months just to stand still. That single line says more about the state of the AI race than any model release could. We are no longer in a software cycle. We are in an industrial one, where progress depends on data centres, silicon supply, energy availability, physical footprint, and geopolitical access.

You could see that geopolitical undercurrent everywhere this week.

The UAE’s decision to invest $1 billion into African AI infrastructure is not a regional experiment, it is a strategic expansion of influence through compute. Data centres are becoming diplomatic instruments. Sovereign infrastructure is becoming soft power.

Markets echoed this momentum. Nvidia delivered another strong quarter, easing concerns of an AI slowdown while intensifying questions about global dependence on a single hardware backbone. When one company becomes the proxy for the world’s AI appetite, you realise this is no longer an industry story, it’s an economic architecture story.

My takeaway from the weekend

Put these threads together and the picture becomes clear. AI has split into two races:

• A capability race, where models like Gemini 3 reset expectations.
• A capacity race, where the world scrambles to build the physical, political, and economic foundations required to run those models at scale.

“The leaders who stay ahead will be the ones who understand that competitive advantage is shifting from “Who has the smartest model?” to “Who can deploy intelligence reliably, responsibly, and at scale?”

Beyond AI: my mindshare – when a Yogi meets an AI pioneer

I came across a fascinating conversation from two iconic leaders;  Spiritual leader and a great Yogi, Sri Sri Ravi Shankar and a leading AI thought leader, Andrew Ng.

One speaks about clarity of mind, the other about clarity of capability. Yet point to the same truth. AI will only move as fast as humans are ready to trust it, understand it, and work with it.

“Sri Sri emphasises inner steadiness. Andrew emphasises skill and confidence. Together, they outline the real leadership agenda: prepare the people as much as the model. Because in the AI age, intelligence is abundant. Readiness is not.”

Weekend Notebook #46 – Gartner IT Symposium Special Edition

amitabh apte

Published on LinkedIn and amitabhapte.com on16th Nov, 2025

This week in AI – Five AI Signals from Barcelona

Barcelona had a different energy this year. The conversation has moved on from what AI can do. That phase is over. The focus now is on how organisations absorb the speed, scale, and structural change AI is introducing.

Across the keynotes, roundtables, and research sessions, five themes kept surfacing, sometimes quietly, sometimes unmistakably.

Theme One: The widening transformation gap.
Some organisations have begun treating AI as infrastructure, with governance, data foundations, ownership, and responsible deployment embedded into normal operations. Others remain stuck in pilot mode, experiments that never scale, uneven adoption, unclear accountability, and a general hesitancy to move. As one analyst put it: AI capability is rising fast, but organisational readiness is not. That tension is now shaping the competitive landscape.

Theme Two: The rise of agentic architecture.
By 2028, most B2B buying will be mediated by AI agents, and most customer processes will be handled by multiagent systems. This is not workflow automation. It is workflow replacement. Processes that once followed structured steps are becoming dynamic, context-aware, and decision-driven. Interfaces are shifting toward conversations. Enterprise platforms, from ERP to CRM, are reorganising around intelligence that acts, not waits.

Theme Three: Governance as the new accelerator.
Not in the traditional compliance sense. Governance has become the mechanism that determines speed. The organisations moving fastest weren’t the ones with the most models; they were the ones with clean data, disciplined model management, strong provenance, clear policies, and embedded risk thinking. In a world shaped by evolving regulation, geopolitical pressure, and rising expectations of trust, governance is no longer the brake, it is the runway.

Theme Four: Real value from AI.
This was the moment the conversations got honest. Leaders are discovering that “time saved” does not equal “value created”. Productivity gains are only the beginning. Real value emerges when organisations reengineer processes, redesign decision flows, renegotiate outsourcing, adopt hybrid human–agent operating models, and build AI-native products and services. The organisations making genuine progress are not adding more AI, they are rethinking how the enterprise works.

Theme Five: The shift in leadership expectations.
Technology leadership is expanding from delivery to direction. CIOs, CDOs, and emerging CAIOs are increasingly expected to influence beyond their function, connecting strategy, architecture, operating models, and transformation; converting AI potential into outcomes; and aligning people and processes around new ways of working. Influence, not technical mastery, is becoming the real differentiator.

My takeaway from the weekend

“The next chapter of AI will not be won by those with the most initiatives. It will be won by those with the most coherent organisational design, the strongest architecture, the clearest governance, and the most aligned leadership.”

Beyond AI: my mindshare – Four Human Lessons from Barcelona

Ironically, the most impactful sessions had nothing to do with technology. Four speakers; Bear Grylls, Jo Malone, Chris Barton, and Charles Duhigg, offered a blueprint for the human side of transformation, which felt even more relevant amid all the agent diagrams and architecture slides.

Bear Grylls spoke about resilience. Fear is normal. Vulnerability builds trust. Pressure creates capability. And isolation, not danger, is what truly undermines performance. In a world where leaders face ambiguous AI decisions daily, his message felt practical and grounding.

Jo Malone spoke about instinct. Her philosophy was disarmingly simple: notice what others overlook, trust your senses before the data arrives, and treat simplicity as a form of intelligence. AI may amplify creativity, but instinct and taste continue to differentiate great work from average output.

Chris Barton, the founder of Shazam, spoke about perseverance. Shazam was considered impossible for years. Constraints became catalysts. A thousand small iterations produced a breakthrough. His message was clear: as AI automates the easy work, human advantage shifts to originality, first-principles thinking, and unreasonable persistence.

Charles Duhigg delivered a masterclass in communication. Great leaders, he argued, are “supercommunicators”, people who ask deeper questions, match the type of conversation others are having, loop back understanding, and create psychological safety. In an AI-powered world where information is abundant, but alignment is scarce, communication becomes a strategic asset.

“Together, these four voices reveal the traits organisations need most now: resilience, intuition, perseverance, and connection. These are the qualities that anchor teams through uncertainty and accelerate transformation. AI can amplify what we do, but only character determines what we become.”

Weekend Notebook #45 – The Cost of Intelligence

amitabh apte

Published on LinkedIn and amitabhapte.com on 9th Nov, 2025

This week in AI – When Ambition meets Arithmetic

The AI boom is now running on infrastructure, not imagination.

OpenAI’s CFO Sarah Friar spent much of the week clarifying that the company isn’t asking for a government backstop on its $1.4 trillion infrastructure plan, a figure so vast it rivals national energy budgets. Her comments drew a sharp response from Trump’s new AI czar, David Sacks, who declared there would be “no federal bailout for AI.” The exchange revealed the growing tension between private ambition and public patience. The industrialisation of intelligence is proving as capital-intensive as any past revolution, and just as politically fraught.

Yet the capital keeps flowing. OpenAI signed a $38 billion cloud-computing deal with Amazon, making AWS its primary engine for model training and deployment. For Amazon, it’s a strategic coup; for OpenAI, a hedge against the global shortage of compute and chips. The partnership underscores how the AI stack is consolidating, fewer players, bigger bets, tighter dependencies.

Apple talking with Google to power a new Siri using Gemini AI marks a pragmatic turn for the company once obsessed with control. Even Apple is realising that no single firm can build the full AI stack alone. The new race isn’t to own the model; it’s to own the infrastructure, the energy, and the ecosystem.

Tesla shareholders approved a record-breaking $1 trillion pay package, a figure that defies logic until you see the scale of his ambition: turning Tesla from an automaker into an AI-and-robotics platform that spans cars, humanoids, and autonomous fleets. His reward is tied not to quarterly profits, but to an $8.5 trillion valuation. The math may be extraordinary, but it captures the mood of the moment, an economy running on belief as much as balance sheets.

My takeaway this weekend

The age of AI infrastructure is here and it’s expensive.

Across these stories runs the same current: the strain is showing. Building intelligence at planetary scale demands not just algorithms and GPUs, but grids, land, and trillions in capital. The numbers are breath-taking, but so are the risks. The question is no longer can we build it, but how much can the world afford to spend chasing it. Musk’s trillion-dollar vision shows what’s possible when belief meets capital. Friar’s clarification reminds us what happens when optimism meets arithmetic. The future of AI won’t be limited by imagination. It will be priced by reality.

Beyond AI: my mindshare – Human Code Behind Technology Buying

Earlier this week, I joined fellow technology leaders Tom Clark (Everywhen), Rebecca Reynolds Jones (Institute of Directors), and Charlotte Walters (HSBC) on a panel at Computing’s Future of B2B Tech Marketing event, moderated by Computing editor Tom Allen.

Our discussion explored how technology buyers and marketers are navigating an era of automation, AI, and hyper-personalisation and what trust really looks like when so much outreach is now machine-generated.

My core message was simple: credibility begins with relevance.

The best outreach doesn’t just know your name; it understands your business, your pressures, and your purpose. The quickest way to lose a CIO’s trust? Treat technology like a product rather than a partnership.

As AI personalisation scales, human authenticity becomes the real differentiator.
Integrity, empathy, and business fluency still open more doors than algorithms ever will.

“The future of technology marketing won’t belong to those who automate the fastest, but to those who empathise the deepest.”

Weekend Notebook #44 – When the Machine spoke back

amitabh apte

Published on LinkedIn and amitabhapte.com on2nd Nov, 2025

This week in AI – The Infrastructure Age Deepens

The AI boom is no longer about smarter algorithms; it’s about the race to build the physical backbone of intelligence.

Nvidia has now passed $5 trillion in market value, while OpenAI is reportedly preparing a $1 trillion IPO. Together they symbolize the next industrial age, where compute, capital, and energy converge. Global AI infrastructure spending could hit $3–$4 trillion by 2030, rivalling the scale of past industrial revolutions.

This wave extends far beyond tech. Data centres are the new ports, energy grids the new railways, and silicon the new steel. Nations are competing for compute sovereignty; companies are re-architecting their balance sheets around it. The economy of intelligence is becoming tangible, steel, silicon, and power.

Earnings this week underlined the shift.

Amazon’s AWS surged 20% to $33B, reigniting growth and investor confidence. Microsoft’s Azure jumped 40%, even as it warned of record 2026 capex. Alphabet lifted its AI investment to $93B, Meta pledged “hundreds of billions” to build superintelligence, and Apple delivered a record $100B quarter through pricing power and premium design.

Across the board, the message was unmistakable: AI’s centre of gravity has moved from models to machines, from ideas to infrastructure.


This is the industrialisation of intelligence, when data becomes infrastructure, and infrastructure becomes intelligent. When five of the world’s biggest companies post record quarters for the same reason, it’s not a trend, it’s a transformation.

Beyond AI: my mindshare – when the Machine spoke back

It started with a beep.
My washing machine froze mid-cycle, flashing an error code I didn’t understand, then came the smell of something burning.

Normally, I’d begin the usual ritual: search for the paper manual (lost years ago), Google it (wrong model, always!), trawl through forums (half the links broken), and finally land on a YouTube video that spends more time asking me to like and subscribe than showing how to fix it.

This time, I asked ChatGPT.

Within seconds, it decoded the error, explained the cause, asked for a photo to confirm, and calmly walked me through the fix, no pop-ups, no ads, no noise. Just reasoning, relevance, and reassurance.

When the machine started humming again, I realised something had shifted.
For the first time, I hadn’t searched for information, I’d been guided to understanding.


It felt less like fumbling in the dark and more like talking to someone who’d solved this problem a few times before. The internet gave us answers on demand; AI now offers understanding on request.
We once taught machines what to do, now they’re starting to teach us how to think differently.

Weekend Notebook #43 – From Tabs to Tasks: The AI Browser Wars Begin

amitabh apte

Published on LinkedIn and amitabhapte.com on 26th Oct, 2025

This week in AI – The web starts thinking

For three decades, the web browser has been the quiet constant of our digital lives, a passive frame for everything else that changed. This week, that frame began to think.

OpenAI’s new Atlas browser isn’t built for people to browse the web; it’s built for agents to operate within it. Atlas can reason, navigate, and act, booking flights, summarising research, filling forms, all on our behalf. Two days later, Microsoft’s Copilot Mode in Edge arrived, turning every tab into a live workspace that understands context, compares options, and completes tasks without a single click.

The world’s most familiar software is being rewritten around intelligence. If the 2000s belonged to search and the 2010s to apps, the 2020s may belong to interfaces that think. Browsers are no longer windows; they’re companions that interpret intent and orchestrate action.

Behind these headlines runs a deeper current. Microsoft’s autumn Copilot release, with twelve major updates and its own MAI multimodal models, marks a pivot from OpenAI reliance to proprietary reasoning. Anthropic’s million-TPU deal with Google shows that the new scarcity isn’t data but energy. Mondelez, cutting marketing costs through generative workflows, and CrowdStrike, fusing AI with cybersecurity, prove that intelligence is moving from experimentation to execution.

Even hardware is catching the drift: Alibaba’s AI glasses and Amazon’s smart lenses for delivery drivers bring cognition to the edge of human experience, while Google’s quantum breakthrough, computing 13,000 times faster than a supercomputer, points to the next epoch altogether.

Across all of it, a single idea connects the dots, context is the new code.
The most powerful systems will not just know more; they’ll understand where they are, who they serve, and why it matters.

This is not just a redesign of software; it’s the quiet re-architecture of digital work, where intent meets intelligence and the interface becomes the collaborator.

Beyond AI: my mindshare – Jurgen Klopp on Leading with Energy

Leadership, like football, is a game of emotion and endurance. In his conversation with The Diary of a CEO, Jurgen Klopp reflected on self-discipline, love, and drive, the unseen engines behind his years at Liverpool. His philosophy is deceptively simple: “The first person you have to lead is yourself.”

He spoke of turning pressure into purpose and building systems that run on belief as much as strategy. His famous gegenpressing mantra, “If you are not Lionel Messi, you have to defend”, is less about tactics than about trust and accountability.

For leaders navigating transformation, Klopp offers a playbook for the AI age:
lead yourself first, energise others through clarity, and remember that resilience is a renewable resource.

Weekend Notebook #42 – Lights of Progress: from Smart Glasses to Smart Economies

amitabh apte

Published on LinkedIn and AmitabhApte.com on 19th Oct, 2025

This week in AI – When AI Becomes Tangible

This week, AI stepped further out of the cloud and into the real world, shaping markets, moving currencies, and rewriting the geography of innovation.

EssilorLuxottica’s record-breaking quarter sent its shares up nearly 14% to an all-time high. The driver? Its AI-powered Ray-Ban Meta smart glasses. Once a novelty, they now represent a powerful convergence of hardware, intelligence, and design. The blend of form, function, and data is transforming wearables from accessories into interfaces, subtle, seamless, and socially acceptable. When design meets purpose, adoption accelerates.

Meanwhile, OpenAI’s new partnership with Broadcom marks a decisive move from software to silicon. By co-designing custom AI chips, OpenAI aims to reduce dependence on Nvidia and secure its own compute future. This is the next wave of integration, from algorithms to architecture, from models to metal, giving OpenAI control over both intelligence and infrastructure.

Finance, too, is recalibrating around AI’s physical footprint. Goldman Sachs is building a new lending unit to finance AI infrastructure, while BlackRock’s $20 billion acquisition of Aligned Data Centers ranks among the largest in the sector’s history. Infrastructure is now investable; data centres, cooling systems, and energy grids are becoming the new ports and pipelines of the digital age.

The wave is global, and India is fast becoming one of its most ambitious players. Google’s $15 billion investment in a new AI data centre in Visakhapatnam, , underscores India’s “swadeshi tech” ambition to localise AI infrastructure. To power this growth sustainably, the country is also exploring small modular nuclear reactors (SMRs) for round-the-clock clean energy, a bold shift from renewables that opens its nuclear sector to private and foreign investment.

Beyond India, the ripple effects are being felt across economies. Sterling and the Swedish krona are both strengthening as capital flows into new AI data hubs in London and Stockholm. Analysts call it the “compute capital effect”: when technology investment starts to influence currency strength and macro stability. Innovation, in other words, is becoming an economic moat.

And as AI enters new domains, society is adapting in parallel. Instagram’s upcoming parental controls for AI chatbots show how platforms are finally acknowledging their responsibility for young users’ wellbeing. In healthcare, AI is reducing administrative load and clinician burnout, yet only 28% of doctors feel ready to use it effectively. The technology is advancing faster than human capability to absorb it. That readiness gap is emerging as one of the defining leadership challenges of our time.

My Takeaway This Weekend

From eyewear to energy, from silicon to society, one theme connects it all: AI is crossing from the digital layer into the physical economy. It’s no longer something we log into; it’s something we live within.

“This is the industrialisation of intelligence, when data becomes infrastructure, and infrastructure becomes intelligent.”

For leaders, the task ahead is to design for that convergence, where compute, capital, and culture intersect. Because the next decade of AI won’t just be coded in labs; it will be built in factories, financed by markets, powered by clean energy, and worn on faces.

Beyond AI: My mindshare – the Light we share

This week, as millions around the world celebrate Diwali, the festival of lights, homes, offices, and streets glow with lamps, laughter, and the scent of homemade delicacies.
It’s a time to pause, reconnect with family and friends, and celebrate the warmth of togetherness.

For me, Diwali has always been about more than lighting diyas. It’s also a reminder to light the lamp within, the spark of compassion, curiosity, and kindness that brightens the lives of those around us. Each flame we light carries meaning: to share joy, to help someone find their spark, and to bring others along on our journey.

In a world that often feels fast and fragmented, Diwali invites us to slow down and rekindle what truly connects us, gratitude, generosity, and shared light.

“When we light a lamp for someone else, we illuminate our own path too.”