Anthropic turns down investment at a valuation above $800 billion as the AI chip market accelerates

Anthropic has shown that in the new phase of the AI boom, money is no longer the scarcest resource: the company can afford to decline investment offers even at a valuation above $800 billion. Meanwhile, Meta is doubling down on its own chips with Broadcom, and ASML has raised its sales forecast because demand for AI infrastructure keeps pushing the entire semiconductor market higher.

The main news of the three is Anthropic's response to investor interest. The company rejected attempts to invest in it on terms that valued the business at more than $800 billion. That fact matters more than any single funding round: it shows how sharply capital's appetite for developers of foundation AI models has grown. A few years ago, such companies fought for access to compute and funding. Now they can choose not only their partners but also the very moment at which they need outside money at all. The refusal does not necessarily mean Anthropic will not raise capital later, but it clearly underlines its bargaining power.

Investor interest in Anthropic is easy to explain. Companies building large AI models sit at the center of a market where several questions are being settled at once: who controls the best models, who gets them into products fastest, who wins enterprise customers, and who can sustain the pace of spending on training and running them. In such a market, a valuation is not simply a reflection of current financial results but a bet on a future share of a critical technology platform.

If investors are ready to discuss a level above $800 billion, it means the market believes not only in growing demand for generative AI but also in the small number of companies that can realistically claim leadership.

A second storyline is developing in parallel: the fight for compute independence. Meta is expanding its multibillion-dollar partnership with Broadcom to design and produce custom chips. For large platforms, this is no longer a side project but part of core strategy. The deeper AI is built into advertising systems, search, recommendations, enterprise tools and consumer services, the more painful the dependence on general-purpose accelerators and outside suppliers becomes. In-house chips offer a chance to better control the price of compute, tune architecture to specific workloads and reduce the risk of shortages. That matters especially now, when demand for AI capacity is growing faster than the industry can add supply.

The third signal came from ASML, a key supplier of advanced chipmaking equipment. The company raised its full-year sales forecast, pointing to growth fueled by AI demand. For the market this is an important marker, because ASML sits deep at the base of the entire value chain. If model developers and the largest platforms increase their budgets, the next step is orders for new chips, expansion of fab capacity and a growing need for the most complex lithography machines. When optimism rises at such a supplier, it looks less like a local story about one company and more like confirmation that the AI investment cycle is spreading across the whole infrastructure, from software to hardware.

Together, these stories form a coherent picture. The center of gravity in the AI race is shifting from demonstrations of capability to control over the full economics of the stack: capital, compute, supply chain and equipment. Anthropic shows that strong model developers can dictate terms to investors.

Meta shows that the next level of competition is in-house hardware and a more efficient cost of running AI services. ASML confirms that the boom is not limited to apps and chatbots but is already materializing in orders for the semiconductor industry. For the market this means one simple thing: the winners of the AI cycle will be determined not only by the quality of their models, but by their ability to control the resources those models run on.

*Meta is designated an extremist organization and banned in Russia.