Lightspeed, an early investor in Anthropic: Why can Anthropic win?

CN
链捕手
Follow
2 hours ago

Authors: Ravi Mhatre, Guru Chahal, Sebastian Duesterhoeft, Lightspeed investors

Compiled by: Gu Yu, ChainCatcher


In the early days of Anthropic, 15 people gathered weekly at Precita Park in San Francisco for lunch and strategic meetings. It was 2021, and the pandemic still loomed over the city. They brought kitchen chairs from home, spread out on the grass, and discussed the future of intelligence.

They aimed to create the most capable models in the world. They also wanted to ensure that humans could trust it.

At that time, there was no product, no business, and no one knew how rapidly science would advance. But the people in the park had been working at the forefront of artificial intelligence for years. They knew where the technology was headed and gradually believed that its true potential could only be realized if it was understood and controlled with the same seriousness of commitment.

This belief became Anthropic.

In the following years, the pace of the company's progress almost exceeded all our expectations. Capabilities once relegated to science fiction became reality. Throughout it all, Anthropic maintained a meaningful commitment: to build intelligence that is both trustworthy and safe.


A Belief That Precedes the Company's Existence

The origins of Anthropic go back much further than those afternoons at Precita Park. In June 2016, Dario Amodei, Chris Olah, and four other researchers published "Concrete Problems in AI Safety." While most of the world was still debating whether truly powerful AI would ever arrive, these researchers were already asking: What will happen when it arrives? This paper opened new avenues and established AI safety as an engineering problem by posing concrete, testable questions that could be studied and addressed before the risks became high.

As Dario and Chris advanced in their careers, this steadfast commitment to building safe AI systems continued, reaching its peak with the birth of Anthropic: Daniela Amodei, Jared Kaplan, Tom Brown, Sam McCandlish, and Jack Clark joined them, collectively upholding this sense of responsibility.

When we first met Dario and the team, there was no evident business model. Dario admitted that in those early days, they knew very little about how the company would make money. However, two things were clear. First, AI was about to become one of the most powerful forces in human history. Second, those who built it had to embed safety into the foundation, rather than adding it on afterward.


When Intelligence is No Longer Scarce

Every great technological revolution has expanded what people are capable of. The printing press made knowledge reproducible, electricity changed the physical limits of labor, and computing enabled us to process information at a massive scale. Artificial intelligence is different—it extends precisely the very capability we use to discover, invent, and achieve progress.

Anthropic is helping to build a future where intelligence is abundant.

In such a world, scientists can test hypotheses at scale and easily draw on other disciplines. In that future, new drugs will reach the market, and people will approach work in innovative ways, with everything from transportation to urban development being redesigned.

This grand ambition may feel overwhelming. If Anthropic's vision for the future comes true, decades of scientific progress could be compressed into just a few years. Our lives could look and feel vastly different from today.

Fifty years from now, when we look back at this moment, we may wonder how we lived through it.

Science has always been an open question. From the early Greek stargazers who questioned the universe to generations pushing the frontiers of physics, biology, mathematics, and chemistry, we are entering the next major turning point for humanity, and the Anthropic team is providing us with the tools to push intelligence to new heights.

Thus, the issue of trust is crucial. One of the most common questions colleagues and friends ask us is: How seriously should we take forecasts about the future, and who should we rely on in this moment?

For us, Anthropic’s openness—whether it’s Claude’s Constitution, Responsible Scaling Policies, or its attitude towards controlling the pace at the frontier—indicates this is a company that takes its responsibilities seriously and always welcomes scrutiny and debate. This is not only a key element of its success but also the reason it has earned the trust of so many.


People Behind the Mission

A significant part of Dario Amodei’s career has been spent contemplating the future of intelligence. He combines scientific rigor with a rare ability: the insight to see where technology is headed long before others do. In just a few years, he transitioned from researcher to founder, becoming the leader of an organization with global impact. Most founders, even if they can make these transitions, take decades; Dario achieved it in just a few years.

Yet he has remained curious, intellectually honest, and willing to revisit his conclusions. He thinks about leadership from first principles, as he does with research, and regards leading Anthropic as a form of service.

Daniela Amodei is equally indispensable. Her understanding of the risks of hypergrowth is unparalleled. The pace of an organization's expansion can outstrip the speed at which its culture can absorb it, and when hundreds become thousands, maintaining the trust of a small founding team becomes difficult. She has consistently and consciously protected that trust.

All seven founders—Chris Olah, Daniela Amodei, Dario Amodei, Jack Clark, Jared Kaplan, Sam McCandlish, Tom Brown—remain at Anthropic, demonstrating the culture the company has built. At the forefront of AI, some of the most significant decisions have no precedent. In those moments, the company needs people it can trust enough to voice dissent, reason together, and return to shared principles. The founders built this trust over years of tackling challenges, and it has become one of the company’s greatest assets.

The same qualities extend beyond the founding team, with Krishna Rao being the clearest example.

The funding needs to build frontier intelligence differ almost entirely from any company in history. The required capital and computational resources are immense, the partnerships complex, and decisions come quickly. At the forefront of AI, the ability to obtain capital and computing power is intimately tied to the ability to advance science. Krishna has provided Anthropic with the resources and endurance necessary to pursue its ambitions, continually finding ways to make what once seemed unattainable possible.

His contributions are foundational to the company’s current achievements.


A Steadfast Mission

When a company achieves such success, people often recount its story in reverse. The scale of the business makes results seem inevitable, while uncertainty and courage fade into the background of everything that follows.

The emergence of Anthropic was by no means guaranteed.

The founders left stable positions to start anew, driven by scientific beliefs and a profound sense of responsibility for the future they could foresee and that was coming. They did not know if they could build a company on this belief. They did know why they had to try.

Everything they built from then on changed the world’s belief in what is possible. A future that once belonged only to imagination is taking shape around us, and with every advancement, the questions multiply. The founders have always believed that the benefits must broadly accrue to humanity, and that the institutions creating this technology must earn the trust of the societies they impact.

This is why the character of Anthropic's founders is so important to us. We have gradually come to know a group of exceptionally capable people who understand that their greatest responsibility is to ensure that the progress of intelligence lives up to human trust.

If we must choose such people, we would choose them.

The mission has never wavered. Its most important work still lies ahead. The world has become keenly aware of the possibilities that intelligence opens up.

免责声明:本文章仅代表作者个人观点,不代表本平台的立场和观点。本文章仅供信息分享,不构成对任何人的任何投资建议。用户与作者之间的任何争议,与本平台无关。如网页中刊载的文章或图片涉及侵权,请提供相关的权利证明和身份证明发送邮件到[email protected],本平台相关工作人员将会进行核查。

Share To
APP

X

Telegram

Facebook

Reddit

CopyLink