Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
About the Author
About the Author
About the Author

The Jedi Behind the Force of AI: As the Dark Side Creeps, Do We Need AI to Go Nuclear—Not Just for Power, but for Restraint?

Being Human for a Better Tomorrow in the Age of AI

  • November 7, 2025
  • 60th Edition
1762517471932

The 60th Milestone: A Full Cycle and a Critical Choice

In mathematics, 60 is a highly composite number, rich with divisors. In the Indic ethos, it marks a Samvatsara—a full cycle of time, signifying maturity, reflection, and a new beginning.

Reaching this 60th edition mirrors my own state of mind. As an AI user, observer, and advisor, through the lens of AI allyship I see how the skilled use of this technology acts as a tremendous force multiplier, elevating human agency and expression. The builders, businesses, and users have much to be excited about.

Yet, precisely because of this power, I also see a creeping dark side—perhaps not a sudden catastrophe (God forbid), but a slow erosion visible in socio-economic indicators and human impact.

This edition culminates the third part of our recent civilisational vistas:

In the 58th edition, we saw how AI systems are extensions of human minds, shaped by the distinct traits of their creators.

In the 59th edition, as Physical AI emerged, we explored the responsibility of “parenting models” these new intelligences through the lens of fiction.

Now, the 60th turns from creators to the Believers and Enablers (BEs). More than a label, BEs are an existential metaphor: the users, investors, and policymakers who will decide what AI becomes—and what our civilisation becomes with it.

AI, like nuclear energy, holds a dual potential for boundless transformation and utter destruction. Nuclear achieved a fragile but enduring restraint through specific global mechanisms.

Can AI build restraint without a catastrophe—or will it take a Hiroshima-like shock to find our will?

This milestone edition is an exploration of the “Jedi” working for that restraint, the creeping “Dark Side” from reckless use and how we might harness this great power to flourish rather than diminish or perish.

The AI Jedi: Believers and Enablers (BEs) of a Different Kind

In the grand narrative of technology, we often cast the breathless accelerator as the hero.

But in the Age of Intelligence, a different kind of hero is emerging. These are the true Believers and Enablers (BEs)—those who believe so deeply in AI’s potential for human flourishing that they are willing to fight for the restraint necessary to ensure it.

These BEs don’t just want to build AI; they want to build Beneficial AI. They use their agency not just to speed things up, but to steer them right.

We see this drive in the scientists who built the foundations and are now actively redirecting them like Turing Award winners like Geoffrey Hinton and Yoshua Bengio aren’t trying to kill AI; they are calling for a massive reallocation of R&D—one-third of budgets—toward safety and alignment (Science, May 2024).

They are enabling a future where humanity survives its own invention. They are risking their legacies, moving from being celebrated fathers of the field to its most prominent conscience, enduring backlash from accelerationists to call for a massive reallocation of R&D toward safety.

We see it in regulators who are racing not to stifle innovation, but to create the tracks it can safely run on. The EU AI Act isn’t just a red light; it’s a sophisticated signalling system that bans unacceptable risks while demanding transparency for powerful models, ensuring innovation doesn’t come at the cost of fundamental rights (European Parliament, June 2024). This is not easy.

Regulators face pressure from trillion-dollar interests, yet are staking careers on building tracks for AI to run safely, not derail.

We see it in a different breed of capitalists. Investors like Jaan Tallinn are directing capital through vehicles like the Survival and Flourishing Fund solely into AI alignment and safety organisations, prioritising long-term civilisational survival over short-term exits.

And we see it in the open-science community, the unsung heroes of transparency.

Platforms like Hugging Face have become essential for transparent model hosting, while rapid-publishing archives like arXiv allow a global community to spot flaws, test for bias, and red-team models faster than any closed corporate lab could.

By organising through consortiums like MLCommons to create independent safety benchmarks, they are democratising not just access to AI, but the ability to audit it.
What unites these diverse BEs isn’t a policy paper, but by and large a shared set of human virtues—a pattern we can all adopt:

  • Stewardship: They value the long-term safety of civilisation over the immediate dopamine hit of the next breakthrough.
  • Service: They ensure the immense gains of AI serve the public good, rather than privatising profits while socialising risks.
  • Integrity: They practice radical transparency, demanding we clearly label what is human and what is synthetic, refusing to hide behind algorithmic black boxes.
  • Discipline: They have the power to deploy anything, but the wisdom to pre-commit to restraint—pausing if safety thresholds aren't met.

The Creeping Dark Side: A Fuse Burning Toward Degeneracy

While the Jedi fight for restraint, a darker reality is already taking hold. It may look like a slow corrosion today, but it is effectively a long fuse burning toward degeneracies—economic collapse, political chaos, or even an existential threat.

We are not just waiting for a catastrophe; we are actively funding its precursors.

The Socio-Economic Fracture

The harms are no longer theoretical; they are statistical and severe.

The Displacement Crisis

The “automation squeeze” is here. The latest Challenger and Gray report shows 150K jobs decline in US highest since 2003 (Challenger, Gray & Christmas, 2025), some of it is attributed to AI. AI is impacting across countries. Gen Z is facing a brutal unemployment rate and slump in entry-level tech hiring. 41% of 2025 graduates and 25% of 2024 batch in the US are yet to find a job (National Association of Colleges and Employers, 2025). In the UK 1.6 million graduates compete for 17000 jobs (UK Office for National Statistics, 2025). This isn’t just a market cycle; it’s the early signal of structural jobless growth that could trigger widespread social unrest.

Industrial-Scale Theft

The economic foundation of human creativity is being strip-mined. With over 30 major copyright lawsuits active (Copyright Alliance, 2024) and the US Copyright Office flagging training on human work as prima facie infringement (US Copyright Office, May 2025), we are witnessing the largest unauthorized wealth transfer in history.

Physical Debt

The “intangible” cloud has a devastating physical toll. Data centers are on track to consume on some forecasts, 8-10% of U.S. electricity and billions of cubic meters of water annually by 2030 (Pew Research Center, October 2025; Lawfare, December 2024), setting the stage for direct conflict between AI compute and human survival needs.

The Cognitive Collapse

Beyond economics, our shared reality is dissolving.

  • Toxic Information: Our ecosystem is being poisoned by "AI slop"—low-quality, machine-generated noise that now accounts for 15% of top news traffic (NewsGuard, 2024).
  • Weaponised Reality: Deepfake harassment incidents have surged 310% year-over-year (Sensity AI, 2024). We are one sophisticated deepfake away from a triggered political crisis or financial panic that no swift correction can undo.
  • The Atrophy of Agency: Perhaps most dangerously, we are outsourcing thought itself. We prompt instead of write; we summarise instead of analyse. This "Responsibility Laundering"—where humans offload difficult decisions to opaque "black boxes"—isn't just a change in workflow; it is a civilisational de-skilling that leaves us vulnerable to an intelligent artificial being we may one day no longer understand or control.

This is the dark side: distributed today, potentially cataclysmic tomorrow. It is quietly rewriting the terms of our existence while we scroll, waiting for the moment the cumulative corrosion causes the structure to collapse.

Nuclear Energy and AI: Hiroshima Moment or Hindsight

As we saw, AI is insatiably power-hungry, and nuclear power is ironically seen as one of the few scalable, clean sources to feed it.

But going beyond seeing nuclear energy as a power source, they share similar traits. Both are dual-use technologies that, on one hand, can transform human society and, on the other, destroy life.

Yes, there are critical differences. Nuclear governance relies on controlling a physically scarce and resource-intensive material. It’s a play with few actors—governments and multilateral agencies like the IAEA, which conducts thousands of inspections annually to verify compliance. AI, in comparison, is a “Wild West.” Its “material”—code and models—thrives on scale, is infinitely reproducible, and pervades everywhere.

But this was not so in 1945. Nuclear restraint then was not a precaution; it was a reaction.

It took the horror of Hiroshima and a subsequent arms race to set off alarms and force global consensus. It took 12 agonising years to establish the IAEA (1957) and 23 years for the Non-Proliferation Treaty (NPT) to enter force (IAEA, 2005; United Nations, 2017).

Despite the stockpiles, this framework has largely worked. We have thankfully avoided nuclear war for over 70 years, thanks to a collective, generational realisation of its horrors.

When humanity last faced a technology that could both illuminate and annihilate, it learned—through pain—how to govern power.

Can we learn from nuclear’s beneficial use and safeguards, or must AI have its own “Hiroshima Moment” first?

That moment may not be a concentrated bomb but a catastrophic corrosion of lives through:

  • Economic and social fracture from years of jobless growth.
  • The collapse of original Art that celebrates the human mind.
  • Political tensions and the erosion of shared reality, triggered by deepfakes and AI slop.

The Nuclear Lesson: An Inspiration, Not a Blueprint

So, what can we learn from the nuclear age? We learned, the hard way, to build mechanisms for restraint.

We created the International Atomic Energy Agency (IAEA) as a global watchdog, with a mandate for inspections and compliance. We forged international consensus through the Non-Proliferation Treaty (NPT), a framework to prevent the spread of weapons. And we saw key enablers—scientists like Oppenheimer and Einstein—use their moral authority to advocate for this global governance.

But this restraint was bought after a catastrophe and built agonizingly slowly. It took 12 years after Hiroshima for the IAEA to be established and 23 years for the NPT to enter force.

We do not have the luxury of waiting for an “AI Hiroshima” to start the clock. We cannot afford to build our AI safety regime in 2050. The nuclear playbook, which relies on controlling scarce atoms, is not a blueprint for controlling abundant bits.

The true nuclear lesson is not structural; it is inspirational. It is the only time in history our civilization successfully governed a dual-use, existential technology. It’s proof that global restraint is possible, even if the “how” must be different.

The Civilizational Choice: Which BE Will You Be?

This is the fundamental difference, and our big hope.

The nuclear BEs of 1945 were reactive. They were forced to build a consensus from the rubble of a preventable disaster.

The AI “Jedi” like we identified earlier and many such more beneficial AI users are proactive. Our modern Oppenheimers and Einsteins are shouting their warnings before the bomb drops, not after.

The choice is not whether to have believers and enablers. The choice is which kind we will be.

Will we be the reactive BEs of 1945, who only act once the catastrophe is undeniable?

Or will we be the proactive BEs of 2025—the kind that ‘sees’ the dark side creep and has the wisdom to build restraint before it is too late.

It is BEing human in Beneficial AI for a better tomorrow for all.

What do you think?

  • Given the 'Jedi' are already building the immune system for AI, which of their virtues—Stewardship, Service, Integrity, or Discipline—will you embody in your own use of this technology?
  • Having seen the evidence of the 'Creep,' what data points on the dark side trajectory convinced you this is a systemic problem, not an isolated issue?
  • The nuclear analogy shows our governance tools are outdated. What does 'restraint' look like in your domain—your company, your policy, your research—when you can't put the genie back in the bottle?

Thanks for reading and sharing. Welcome your support and look forward to your reactions and thoughts.

Best wishes.

“AI is a language. Treat it like one: practice, iterate, and mind your grammar prompts, assumptions, and verification.”

II. Questioning / Asking

Good conversations flow from well‑sequenced questions—topical, simple, coherent, cohesive.
LLM Conversation Example 1
Q: What are empirical judgments?

A: Empirical judgments are based on observation, experience, or experimentation.
Q: What are moral judgments?

A: Moral judgments are based on ethical principles and values.

IV. Judgment

“The structure of a language affects its speakers’ worldview and cognition.”
— Henry Hazlitt
“The art of questioning is the source of all knowledge.”
— Thomas Berger
Share:

2 Responses

  2. Thank you for your sharing. I am worried that I lack creative ideas. It is your article that makes me full of hope. Thank you. But, I have a question, can you help me?

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *