📡 Thoughts on Uploaded Intelligence after binge-watching Pantheon

I’ve been deep in Pantheon lately, and it’s genuinely rewired how I think about the future. While the world is obsessed with Artificial Intelligence right now, we’re seeing this wild re-genesis of Neural Computing (neuromorphic chips, spiking networks, the whole “brain-like” hardware revival under fresh branding). The show keeps pulling me toward the next frontier: Uploaded Intelligence (UI).

Mind uploading. A full human consciousness, scanned and running in silicon.

Storage-wise? We’re already good. The human brain has roughly 86 billion neurons and ~100 trillion synapses. High-resolution connectome mapping (the wiring diagram) is estimated at a few petabytes for the static structure. Totally manageable on today’s data-center scale. Even with dynamic state (all the electrochemical chatter), we’re talking exabytes at worst. Disk space isn’t the bottleneck.

The real question is compute.

What does it actually take to run a brain in real time? The biological version sips just ~20 watts while delivering insane parallelism. Replicating that neural network density in hardware, not just abstract deep-learning layers, but something that faithfully mirrors the brain’s messy, recurrent, spiking architecture, is a whole different game.

If we could build a system with comparable synaptic density and fidelity, could we then upload the actual synaptic firing patterns? The dynamic “software” state that is you at any given moment?

And here’s the million-dollar (or trillion-dollar) question: How the hell do we perform the upload?

In Pantheon they go the dramatic route: a destructive “Deep Scan” that lasers the brain layer by layer while building the digital twin. Real-world connectomics is already doing this at small scale (they’ve fully mapped the fruit fly brain), but scaling to humans is still sci-fi territory. Non-destructive options (nanobots swimming through living tissue, ultra-high-res MRI on steroids, etc.) feel even further out.

So… how far are we? Decades? Centuries? Or is whole-brain emulation quietly creeping up on us while we argue about chatbot ethics?

Would you want to upload? Would you trust the copy to still be “you”? And if UI becomes possible, does it make today’s AI look like finger-painting by comparison?

I’m genuinely curious where this goes. 🔥

I attended a really interesting panel discussion last week.

Honestly, I was slightly terrified at first - the ratio of lawyers in the room was high enough to warrant a "without prejudice" disclaimer 😂

Jokes aside, I thoroughly enjoyed it. We dove into accountability, impact, policies, and good governance - all wrapped around the big question: how do we actually get sh!t done?

One line that really stuck with me was:

"What if your company is protecting the very rules that are holding it back?"

That question sent me down a rabbit hole… Are we unconsciously forcing non-determinism to behave like determinism?

The Sacred Machine: Non-Deterministic AI in a Deterministic World

🔥 The Making of a CDO and CAIO

Over the past few years, many professionals have asked me: What does it really take to become a Chief Data Officer (CDO) or Chief AI Officer (CAIO)?

There’s no simple linear path. Both roles are shaped by context, delivery, and credibility. Here’s my perspective:

The CDO Role
The CDO has evolved from a visionary data leader into a regulator-mandated guardian of data quality, governance, privacy, and compliance.

It demands deep expertise in data architecture, master data management, regulatory frameworks, and enterprise platforms. Regulators hold CDOs personally accountable. This is not a role for generalists.

Typical path: Progressive leadership in data engineering/analytics, turnaround experience, cross-functional exposure (IT, risk, finance), and proven outcomes like cleaner data and reduced risk.

Market trends show CDOs shifting toward decision enablement and AI readiness, becoming true strategic partners.

The CAIO Role
The CAIO is newer and still evolving. Historically reporting to various leaders, the role is now increasingly reporting directly to the CEO to drive real change.

The mandate sounds simple: identify AI opportunities and manage risks. But it’s actually a demanding transformation and change management role. It requires scaling adoption, building capabilities, and delivering measurable business impact.

Diverse backgrounds succeed here: from domain experts with little prior AI experience to technical specialists. The key? Bridging strategy, technology, and execution.

Market reality (IBM 2026 study): 76% of organisations now have a CAIO, up from just 26% in 2025. Companies with dedicated CAIOs see around 5% higher returns on AI investments.

What It Really Takes
Neither role is for everyone. You need:
- Credibility through real delivery
- Comfort with accountability
- Systems thinking + business acumen
- Relentless learning

Focus on leading major transformations, owning business outcomes, and building executive presence. The path is rarely linear. It’s forged in messy data realities and organisational change. 🚀

Don't use AI just for the sake of using AI

Happy birthday to all May babies 🎂 and some unsolicited reflections 😁

https://www.linkedin.com/posts/davidrh_tomorrow-is-my-birthday-and-im-spending-activity-7466412191164911616-5jqY?utm_source=share&utm_medium=member_android&rcm=ACoAAAC11PUB6bXVB2Khjqo0ABlxtMcjJuzhmtw

Progressively available :) In case, for whatever reason, you actually end up reading it - hope it makes you laugh.

https://www.linkedin.com/posts/davidrh_aifailure-responsibleai-digitaltransformation-share-7466950656164302848-WfJx

🚀 The AI Layoff Mirage: Why Headlines Ignore the Jobs Being Created

Tomorrow, I’m publishing a new opinion piece that challenges the dominant narrative around AI and jobs.

Headlines are filled with stories of layoffs at Meta, Oracle, Shopee, and other tech giants, with AI being blamed for nearly 40% of recent job cuts. It feels alarming — but this is only half the story.

In reality, companies are not destroying jobs indiscriminately. They are pruning inefficiencies and redirecting talent toward exciting new opportunities in data centers, energy infrastructure, finance (AI oversight, personalised advisory), healthcare, and many other sectors.

The piece explores why the labor market remains resilient, what successful adaptation looks like, and why it’s time to move past the fear-mongering.

Stay tuned — the future is brighter than the headlines suggest.

#AI #FutureOfWork #JobMarket

The AI Layoff Mirage
https://www.linkedin.com/pulse/ai-layoff-mirage-dr-david-r-hardoon-umjsc

Thank you!! 😁

https://www.linkedin.com/posts/davidrh_aifailurehandbook-ai-enterpriseai-ugcPost-7472094009704833024-qBX5/

“If I had an hour to solve a problem and my life depended on the solution, I would spend the first 55 minutes determining the proper question to ask, for once I know the proper question, I could solve the problem in less than five minutes.” — Albert Einstein

Why “What is the AI ROI?” Is Usually the Wrong Question and How to Ask the Right One Instead

https://www.linkedin.com/pulse/why-what-ai-roi-usually-wrong-question-how-ask-right-one-hardoon-i9l4c?utm_source=share&utm_medium=member_android&utm_campaign=share_via

Hey everyone 👋

Haven’t been posting much lately — been getting back on the corporate bicycle after some time away. But I’m back with more content, thoughts, and views, starting with this one.

I’ve written before about why HR needs transformation, and this is a direct continuation of that. AI-first never meant human-second, and it was never meant to be a false choice between AI or human.

Is AI Really To Blame for Layoffs?

Had the pleasure of joining Charmaine Koh, Xueling Tan, Patrick Tay, and Ian Yong Hoe Tan on Talking Point with Diana Ser to tackle one of the hottest questions right now.

My biggest takeaway (and the line I keep repeating):

AI-first never meant Human-second.

AI is absolutely reshaping work, but the best organisations and leaders are using it to amplify human capability, not replace it. Judgment, empathy, creativity, adaptability, and the ability to reimagine roles — these deeply human strengths remain the real differentiators.

Grateful to be part of a thoughtful discussion that included real experiences from retrenchment, union perspectives, and industry insights.

Watch the full episode here:
https://www.mewatch.sg/watch/Talking-Point-2026-2027-E10-Recent-Layoffs-Is-AI-Really-To-Blame-819503

What are your thoughts. Is AI being unfairly scapegoated for layoffs, or is there more to it in your industry?

#AI #FutureOfWork #HumanCenteredAI #WorkforceTransformation #Reskilling

Completely random 🌟🎶 I love music, and i'm still one of those who enjoys listening to Radio, on walks, on road trips and in new countries. but - I not a fan of listening to ads or talking. Then you need to figure out which radio station to jump to next.

So - I've build an Android App (still figuring out the name) that does it for you. One can add ANY radio station worldwide, whether based on search, or proximity. Order them by preference. The app will automatically detect ads or talking and will skip to the next station.

Thinking of whether to release this on Google PlayStore 🤣

Across boardrooms, hospitals, banks and government offices, a quiet madness has taken hold.

Leaders claim revolutionary gains while projects fail, chatbots go unused and honest questions become dangerous. Dissenters are sidelined. True believers rise.

This is not hype. It is institutional psychosis where no one can safely speak the truth.

From the front lines of consulting, one observer has watched every AI initiative fail for over a year. The emperor has no clothes, yet the march continues.

“AI Mania Is Eviscerating Global Decision-Making.”
A clear look at the cult, the silence, and how the sane survive.


https://hermit-tech.com/blog/ai-mania-is-eviscerating-global-decisionmaking

I just got rejected by arXiv. 📨

The moderators said the work lacked sufficient original research and wasn’t of interest.

I never planned to submit it in the first place.

It made me think about how ideas now reach people, and how often the audience arrives before the gatekeepers. 👀

I wrote a short piece on it. ✍️

https://www.linkedin.com/pulse/when-audience-arrives-before-gatekeepers-dr-david-r-hardoon-0wrhc

A left field topic to start the week. What do you think?

https://www.linkedin.com/pulse/money-cant-buy-you-love-babies-dr-david-r-hardoon-uhhsc?utm_source=share&utm_medium=member_android&utm_campaign=share_via

A few weeks ago an intern pulled me aside, fairly concerned.

I had said that each of us can now go to a depth, and a breadth, that was simply not possible before. She heard that as a claim about expertise. It is not. It is a claim about the cost of finding out.

My phone had four bars and nothing worked. The ordinary responses were all available to me. Call the operator. Try a few settings. Change operator. I did none of those. I built a tool, with no experience of ever having built an app, that tells me why.

Ten days of readings, one city, one network. One mast put 84% of its readings below the level where calls and video stop working. Another, a few streets away, managed 3%. Same phone, same operator, same afternoon. One mast was 97% unusable at 11am and 2% unusable that evening. Not a bad mast. A bad mast at a bad hour.

The part that stopped me: those first two masts read identical signal strength. Both −91 dBm. Four bars each. The number the phone shows you cannot tell them apart, and the number that can — how clean the signal is, not how loud — it never shows you at all.

That used to be a conversation you had with a network engineer, if you could get one. The bars are raw power against cut-points the carrier chose. They say nothing about a cell full of other people’s traffic, and nothing about the interference on top of you. Once signal quality drops below zero, connection tests fail three times as often and the slow ones take three times as long. Measurable on any handset. Displayed on none.

The app cannot choose a different mast. No app can. The original aim was to make the phone heal itself, and that is not possible. What it can do is end the argument about whether it is your phone, your plan, or you.

An application for a user of one. That is the point she was worried about, and the point I was making. The entry cost is now low enough that going and looking is no longer reserved for the people whose job it is.

https://play.google.com/store/apps/details?id=com.signalscope&pli=1