Anthropic, the artificial intelligence company founded by former OpenAI executives, is reportedly nearing a $1 trillion valuation as it prepares for a US public listing. The news has sent ripples through global markets and placed the UK’s Competition and Markets Authority (CMA) on high alert, underscoring the mounting regulatory scrutiny surrounding frontier AI firms.
The San Francisco-based startup, known for its safety-focused Claude AI model, has been in talks with investment banks for a potential IPO that could value it at over $900 billion, with sources close to the matter suggesting the $1 trillion mark is within reach. This would place Anthropic in the same league as tech titans like Apple, Microsoft, and Nvidia, a remarkable feat for a company founded just four years ago.
Anthropic’s meteoric rise reflects the insatiable demand for generative AI technologies. Its flagship product, Claude, competes directly with OpenAI’s ChatGPT and Google’s Gemini, but with a distinct emphasis on ethical safeguards and interpretability. The company’s “constitutional AI” approach, which aligns model behaviour with human values, has won plaudits from policymakers and researchers alike. Yet this very success has triggered alarm bells at the CMA, which fears that an Anthropic listing could concentrate market power in an already oligopolistic sector.
The CMA’s concern stems from Anthropic’s deep ties with Amazon and Google. Amazon has invested $4 billion in the startup, while Google holds a 10% stake through a $2 billion convertible note. These alliances, the regulator argues, could lead to “interlocking directorates” and reduced competition. The CMA launched a preliminary probe in 2024 into Amazon and Google’s partnership with Anthropic, focusing on whether the deals stifle innovation in the UK’s nascent AI market. Although that investigation closed without formal action, the looming IPO has revived fears of vertical integration.
From a Silicon Valley perspective, Anthropic’s trajectory is a double-edged sword. On one hand, it validates the thesis that safety-first AI can be commercially viable. On the other, a $1 trillion valuation raises uncomfortable questions about valuation metrics. The company’s annualised revenue is estimated at just $500 million, giving it a price-to-sales ratio of 2,000 times. Such exuberance echoes the dot-com bubble, albeit with arguably more transformative technology underpinning it.
The listing also highlights a broader geopolitical dynamic. The US continues to dominate frontier AI development, with Europe and the UK struggling to retain talent and capital. The CMA’s interventionism, while well-intentioned, risks driving AI companies away from London’s capital markets. British financial institutions have already lost the IPO of Arm Holdings to New York; another defection would be a blow to Chancellor Rachel Reeves’ vision of a “tech-forward” UK economy.
Yet the regulator’s vigilance is not without reason. Anthropic’s technology, while impressive, harbours existential risks. The company’s own research suggests that advanced AI systems could eventually evade human control, a scenario dubbed the “alignment problem.” The UK’s AI Safety Institute, funded by the government, has been collaborating with Anthropic on robustness testing. But with a trillion-dollar market cap, the incentives for profit could overshadow safety imperatives.
For the common user, this story is about the future of digital sovereignty. AI is not a abstract concept; it powers the chatbots we use, the recommendations we see, and increasingly, the decisions made on our behalf. If a handful of companies control these systems, they control the narrative. The CMA’s alertness is a defence of democratic accountability. But too much regulation could stifle the very innovation that improves our lives.
As Anthropic approaches this landmark, the window for public oversight is narrowing. The IPO will create a class of shareholders whose primary interest is financial returns, not ethical AI. The challenge for regulators worldwide is to craft rules that protect without hampering progress. It is a high-wire act, with trillion-dollar consequences hanging in the balance.
