Etched Challenges Nvidia: Valuation at $5 Billion

Etched: A Meteoric Rise in the AI Chip Sector

The young American company Etched has recently completed an impressive funding round of $500 million, propelling its valuation to $5 billion. This financial success is accompanied by an equally remarkable announcement: the company has already secured $1 billion in orders for its AI inference systems, even though they are still in the testing phase. This rapid growth illustrates the increasing market interest in alternatives to traditional Nvidia GPUs.

Etched positions itself as a serious competitor to Nvidia in the field of AI accelerators. With this new funding round, the startup has reached a valuation of $5 billion. Simultaneously, it has already secured $1 billion in orders for its first product, which is currently being tested by its initial clients. Investors are increasingly drawn to technologies that promise to reduce inference costs, a major economic challenge for generative AI.

Etched's Strategy to Compete with Nvidia

Founded in 2022 by Gavin Uberti and Robert Wachen, two former Harvard students turned Thiel fellows, Etched takes a distinct approach from the versatile GPUs that dominate the current market. The startup develops specialized chips exclusively for inference, which is the stage where an AI model processes user queries to provide responses.

This segment has become crucial as companies deploy AI models at scale. Inference accounts for the majority of operating costs and often constitutes the main performance bottleneck. Thus, any innovation that reduces energy expenses attracts the attention of investors and major cloud players, as do those that accelerate response times and improve infrastructure efficiency.

After successfully manufacturing its first chip with TSMC earlier this year, Etched is now marketing complete systems, which it refers to as " cutting-edge inference clusters." This offering combines chips, racks, and a software layer to optimize performance. According to Etched, this enables faster, cheaper, and more energy-efficient inference than competing solutions.

Several clients are currently testing these initial systems. Despite this validation phase, the startup already claims to have a backlog of orders reaching $1 billion. This is a strong signal for a company whose first product has not yet been deployed at scale.

An Impressive Valuation That Attracts Investors

With this new funding round, Etched has raised its total funding amount to $800 million. The $500 million round, finalized in December but revealed recently, now values the company at $5 billion.

The operation brought together several institutional investors specializing in quantitative finance and technology, including VentureTech Alliance, Jane Street, Hudson River Trading, Two Sigma, Ribbit Capital, and Stripes, which led the funding. Additionally, influential angel investors in the AI ecosystem, such as Andrej Karpathy, Geoffrey Hinton, Fei-Fei Li, Arthur Mensch, and Scott Wu, along with billionaires Peter Thiel and Stanley Druckenmiller, participated.

This success stands in stark contrast to Etched's situation just two years ago. In 2023, the founders faced numerous rejections from investors, despite a thirty-page document detailing their belief that AI models would soon require specialized chips rather than general-purpose GPUs. The company was then operating with limited resources, on the brink of bankruptcy.

However, the explosion in demand for generative AI is now driving investors to seek the next architectures that enhance performance and reduce operating costs. At the same time, alternative manufacturers are multiplying. Groq recently raised $650 million, while major cloud providers like Amazon, Google, and Microsoft are developing their own AI accelerators. Even OpenAI has committed to this path with its first custom chip designed with Broadcom.

If Etched can deliver on its promises regarding performance and costs, it could help diversify a market currently dominated by Nvidia. The startup could also accelerate the adoption of new architectures dedicated to the most demanding AI workloads.