Silicon Valley's Will to Power: How AI Ethics Became a Global Order

In the last two pieces, we saw how Kierkegaard helps us distinguish ethics from morality, and how Nietzsche unmasks Silicon Valley's AI "ethics" as a priestly project, that is, a moral system presented as universal while serving the interests of those who define them.

But Nietzsche has more to say. Beyond the genealogy of morals lies his idea of the will to power: the insight that life itself tends not just to survive, but to expand, to shape, to impose form. When combined with Heidegger's analysis of modern technology, and with critiques of AI colonialism, this gives us a sharper view of why Silicon Valley's moral authority doesn't just defend itself but relentlessly grows, projecting its order across the globe.

Will to Power in the Age of AI

Nietzsche's dictum is stark: "Life itself is will to power." The drive is not just domination but the expansion of capacity, the refinement of control, the imposition of meaning.

Silicon Valley's moral vocabulary of safety, fairness, alignment, has this quality. Once established, it doesn't sit still. Each principle becomes a platform for further reach. Take OpenAI's Preparedness Framework: by defining "catastrophic misuse" scenarios, it implicitly enshrines the company as the authority to measure and mitigate them. Risk itself is framed in a way that justifies the role of its authors.

For Nietzsche, this is the priestly move perfected: power that presents itself as reluctant, even self-denying, but in fact multiplies its own necessity.

Heidegger's Warning: The World as Standing-Reserve

Where Nietzsche diagnoses the will to power, Heidegger explains its technological medium.

In The Question Concerning Technology, he argued that modern technology doesn't just extend human capacities. It enframes the world (Gestell), revealing people and things as "standing-reserve" (Bestand), resources to be stored, optimised, and deployed.

AI is this logic on steroids. Human language becomes tokenised data. Cultural expression becomes training input. Attention, emotion, even imagination become resources for predictive models.

When Nietzsche's will to power inhabits Heidegger's enframing, the result is not just the growth of institutions but the reshaping of reality itself. What counts as a risk, as alignment, even as "intelligence" is determined within the frame that Silicon Valley builds.

The Colonial Logic of AI

This drive does not stop at California's borders. Scholars like Abeba Birhane call it "algorithmic colonisation": AI systems trained on Western data and categories are exported worldwide as if they were universal.

Facial recognition trained on lighter faces defines what counts as "recognisable."

Content moderation pipelines, based on U.S. legal and cultural norms, decide what billions can say online.

Benchmark datasets built on English internet text become the yardstick of "general" intelligence.

This isn't coercion in the old colonial sense. It's subtler: shaping the categories of recognition, the ontologies of value, the frameworks of thought. As Mohamed, Png, and Isaac argue, even the governance discourse of "safety" and "benefit for humanity" universalises a Silicon Valley worldview while erasing alternatives.

In Nietzsche's terms, this is perspectival morality disguised as universal truth. In Heidegger's terms, it is enframing on a planetary scale.

Why Silicon Valley Can't Stop

Can Silicon Valley restrain itself? Haven't we seen pauses, like the delayed release of GPT-2, or Google's brief moratorium on military contracts?

On the surface, these look like moments of self-control. But philosophy suggests otherwise.

Arthur Schopenhauer, in The World as Will and Representation (1818), describes the will to live as "an endless striving without aim or rest" (Vol. I, §56). The drive itself cannot be suspended; every pause is temporary, every satisfaction fleeting, because striving is the essence of life. Limits, in this view, are not solutions but momentary detours in the same restless trajectory.

Nietzsche rejected Schopenhauer's pessimism, but not the insight into the will's inexhaustibility. For him, the will was not to live but to power: not compulsion, but affirmation. "A living thing seeks above all to discharge its strength, life itself is will to power" (Beyond Good and Evil, §13). Striving doesn't wait for obstacles to vanish; it turns obstacles into opportunities for growth. What Schopenhauer saw as tragic compulsion, Nietzsche saw as vitality: the will cannot stop because stopping would mean ceasing to live.

Read through this lens, Silicon Valley's "pauses" were never genuine renunciations. They were tactical reconfigurations, moments where restraint served the deeper expansion of capacity and legitimacy. Competitive pressure, investor demand, cultural prestige: all these forces conspire to ensure that hesitation becomes acceleration. The institutional will doesn't rest because it cannot rest. To expect otherwise is to expect life to stop living.

The Pattern That Persists

Put these pieces together and the pattern is clear:

Nietzsche. The will to power drives continual expansion.

Heidegger. Technology enframes the world as standing-reserve.

Colonialism critique. This framework is exported as universal.

Schopenhauer. The striving cannot stop of its own accord.

The result is not just an industry defending its position, but a moral and epistemic order that naturalises itself as inevitable.

And that's the real danger: not that AI companies are insincere, but that their sincerity and their expansion are two sides of the same will.

Some pressure may be required...

The will to power does not dissolve itself. Nietzsche is explicit:

Life itself is essentially appropriation, injury, overpowering of what is alien and weaker, suppression, hardness, imposition of one's own forms, incorporation, and at least, at its mildest, exploitation.

Friedrich Nietzsche, Beyond Good and Evil, §259, trans. Kaufmann, 1966

For Nietzsche, power sustains itself through expansion. It cannot voluntarily negate its own drive, even if that drive produces harm. What restrains one will to power is not self-denial but collision with another will.

Silicon Valley's pauses in AI development, OpenAI's delayed release of GPT-2 in 2019 (Radford et al., Language Models are Unsupervised Multitask Learners) or Google's temporary moratorium on military AI contracts after Project Maven in 2018 (Shane & Wakabayashi, New York Times, June 2018), were not real renunciations. They were tactical hesitations, soon followed by renewed acceleration. The will to power turned constraints into momentum.

This is why appeals to corporate self-restraint are structurally inadequate. As Nietzsche would remind us, the will to power converts obstacles into occasions for further mastery. The only meaningful counterbalance must come from external power.

The role of government and regulation

Today, we already see attempts to impose that countervailing force. The European Union's AI Act (European Parliament, 2024) is the world's first comprehensive legal framework for artificial intelligence, categorising systems by risk and banning certain practices outright (e.g. social scoring). It represents precisely the kind of external intervention that Silicon Valley cannot supply from within.

Similarly, in the U.S., the Biden Administration's Executive Order on AI (October 2023) directs federal agencies to set new safety and security standards, with mandatory testing and reporting for frontier models (White House, 2023). While softer than the EU's approach, it signals recognition that leaving governance to corporate self-regulation is untenable.

Scholars echo this necessity. As Abeba Birhane argues, "algorithmic systems extend existing imbalances and exploitations" unless constrained by independent oversight (Birhane, 2020, "Algorithmic Colonization of Africa"). Similarly, Atoosa Kasirzadeh warns that pluralistic-sounding frameworks risk becoming "value-washing" without structural safeguards for real diversity of power (Kasirzadeh, 2024).

In this sense, for Nietzsche, restraint must come from outside because the will to power cannot simply negate itself. In our context, that means democratic states, civil society, and multilateral institutions must act as rival powers pressing back against the self-perpetuating drive of Silicon Valley.

Without this, the industry's trajectory will not be slowed by its own promises. Pauses will continue to collapse into acceleration. Self-proclaimed stewardship will mask deeper consolidation. Only external pressure can reshape the horizon of what AI becomes.

Conclusion: Breaking the Spell

The danger of Silicon Valley's AI project is not that it is uniquely malicious, but that it is uniquely inevitable. Once we recognise it through Nietzsche, Schopenhauer and Heidegger, a sobering picture emerges: a will to power that cannot restrain itself, a technological enframing that reduces the world to standing-reserve, a colonial projection of categories that appear universal, and a regime of power/knowledge that naturalises its own dominance.

This is why corporate "pauses," however well-intentioned, dissolve so quickly. They are not failures of sincerity, but the working-out of the very dynamic that drives the industry. What is presented as stewardship is, in fact, expansion; what is promised as responsibility is, in practice, consolidation.

If there is to be resistance, it must come from the outside. Governments, regulators, and publics must assert themselves as countervailing forces, not to stop technology but to redirect its course. The EU's AI Act and the Biden Administration's Executive Order are imperfect, but they represent the first stirrings of a political will strong enough to confront a technological one.

The question is whether these external pressures can grow fast enough, and bold enough, to prevent a single moral and epistemic order from hardening into the infrastructure of global life.

Because once enframed, the world does not announce itself as colonised. It feels natural. And that is the deepest risk: not domination we can see, but domination we forget is there.

The task, then, is urgent: to ensure that AI does not become another crowd we disappear into, another order of truth we accept without question. If Nietzsche was right, the will to power will not stop itself. If Heidegger was right, enframing will not show its own limits. That leaves the responsibility with us, that is, to confront power with power, and to insist that the horizons of the technological future remain open, contested, and human. The question isn't whether Silicon Valley can stop. It's whether we will stop pretending it can.

Dr Marco Motta works at the intersection of philosophy, data analytics, and AI. He helps organisations understand not just what their data says, but how it should be interpreted and acted on. Drawing on a background in phenomenology and years of experience in analytics consulting, he brings a critical, human-centred perspective to the design and governance of data and AI systems.

A version of this article was first published by Dr Marco Motta on LinkedIn.