Under the rhetoric of existential risk, Anthropic turns market control into a matter of human survival — and profits from it.
There is a delicious irony in how Silicon Valley has discovered itself as a moral philosopher. Anthropic, founded by OpenAI defectors who felt their former colleagues were too cavalier about the apocalypse, recently responded to accusations that it is amassing too much power. The company's defense, according to a report by Wired, is as elegant as it is circular: Anthropic's own commercial success is the necessary condition to ensure that artificial intelligence is developed safely. In other words, trust us, because only we know how dangerous this can be.
The problem with this narrative is not the hypocrisy — in Silicon Valley, hypocrisy is merely the standard currency. The problem is the ingenuity of the mechanism. By wrapping commercial competition and public regulation under the cloak of existential risk, the company executes a brilliant market move. If AI development is, as they suggest, analogous to manipulating pathogens in a lab, then freedom of enterprise is no longer a right, but an unacceptable risk. The logical consequence is that only the right guardians — those with the proper enlightenment and, conveniently, the largest GPU clusters — should be allowed to proceed.
This turns monopoly into a new category. In traditional economics, dominating a market requires crushing competitors through efficiency or acquisition. In the existential risk economy, dominating a market is a moral duty. Critics point to Anthropic's aggressive accumulation of capital, influence, and computational power. The company counters by saying this is, in fact, the face of corporate responsibility. It is a self-validating thesis: the more power we accumulate, the safer you are. If a competitor tries to enter the game, it is not competition; it is civilizational recklessness.
Regulation, in this context, ceases to be a check on corporate power and becomes its accelerator. When a company advocates for the state to establish extremely high barriers for creating AI models, it is not necessarily protecting the public. It is protecting its own position against startups that lack deep pockets to bear the costs of compliance and auditing. The safety debate becomes an invisible moat, dug with the best of declared intentions, but with the practical effect of locking the future of the technology into an oligopoly.
None of these observations mean that the risks of advanced language models are fiction. But we need to be clear about who benefits when panic shapes policy. Anthropic has built a business in which fear is the brand's most valuable asset. If AI is not a ticking time bomb about to detonate, the central premise justifying the company's extraordinary power loses its meaning.
The great trap of the AI era will not necessarily be a machine that decides to eliminate us, but a handful of companies convincing society that only they can save us. When control disguises itself as caution, and monopoly dresses up as ethics, the question that remains is not whether the technology is safe — but whether we are.
Anthropic argues that its own commercial success and accumulation of power are necessary conditions to ensure AI is developed safely, framing its monopoly as a moral duty rather than a standard business strategy.
The author claims that advocating for extremely high regulatory barriers protects established companies from startups lacking deep pockets, effectively using the safety debate as an invisible moat to lock the market into an oligopoly.
The critique is that these companies weaponize fear to consolidate extraordinary power. By convincing society that only they can prevent an AI apocalypse, they disguise market control and monopoly as caution and ethics.