据Seligman Ventures管理合伙人Umesh Padval称,大多数重要的初创企业投资决策往往来自黄仁勋本人。Padval自2020年英伟达以约70亿美元收购网络芯片制造商Mellanox Technologies以来,就在其董事会任职,因此与黄仁勋相识。尽管人们希望黄仁勋的最终继任者能恢复他的交易热潮,但投资者并不乐观认为短期内会发生这种情况。
英伟达最大的押注是支持新云服务提供商、推理服务提供商和开放权重模型,这些对于较小的公司来说成本高昂。“尽管存在这种情况,但黄仁勋的举措是生态系统得以形成的主要原因之一,”硬件公司Ghost联合创始人Zain Javaid说。
“黄仁勋相信开源,坦率地说,他还相信欧洲,”Bankiya谈到黄仁勋为何在全球范围内对AI进行大量投资时表示。这种支持在其他地区如欧洲的停止,“将是一个巨大的资金缺口,实际上也是生态系统缺口,没有人能够填补。”
同样真实的是,英伟达在生态系统中如此主导,以至于有时感觉像是一家垄断企业。通过慷慨支票支持这么多公司,英伟达有效地将这些初创企业锁定在其芯片的使用上,从而帮助巩固了其市场主导地位。
例如,Battery Ventures的普通合伙人Dharmesh Thakker预计,大型科技竞争对手将争夺英伟达的投资王冠,“取决于谁能提供最佳的产品、生态系统和社区组合来竞争CUDA,以及足够的财力来资助生态系统,”他说。CUDA是英伟达的软件平台,使开发者能够在其GPU上运行AI模型。
虽然像Google、Amazon、Microsoft和AMD这样的公司可能会支持成熟的初创企业以促进其芯片和云平台的采用,但Thakker怀疑它们会像英伟达那样向无收入的初创基础设施企业投入数亿美元。
一位风险投资家表示,在黄仁勋领导下,需要1亿美元至5亿美元GPU的基础设施公司通常会从英伟达那里以10亿美元或更高的估值筹集一半的资金,以便起步。芯片制造商的参与让其他投资者更有信心以高估值支持这些公司。
“当英伟达出现在资本表中时,每个人都感到安全,”Padval说。失去这种支持后,投资者表示传统风投可能会在写支票支持这些初创企业前三思而后行,甚至可能暂停某些交易。
虽然所有人都同意黄仁勋对英伟达来说是巨大的损失,但他的一位前员工Amulya Vishwanath,现在经营着自己的VC基金,表示“创始人与领导者必须承担责任”,并且生态系统比一个人更大。“人才将继续蓬勃发展,也必须如此。”
查看原文
It was the resignation heard across the world. Jensen Huang, Nvidia’s embattled CEO who co-founded and led the company for 33 years, had officially stepped down. “It’s like a bomb went off,” one neocloud founder told TechCrunch when he first heard the news.
Over the last several years, Nvidia has been a prolific investor, pouring tens of billions into startups annually and participating in 105 rounds this year alone, often as the largest investor, according to PitchBook. “If that direction changes, it’ll be a big blow to so many young companies,” said Maria Davidson, founder of Relational Capital.
Huang’s absence leaves the startup ecosystem without one of its biggest champions and with far less capital for now, investors say, which could have a massive impact on dealmaking, valuations, funding, and even competition.
“Nvidia is carrying almost every AI-native startup from application to neocloud,” Dawn Capital’s Shamillah Bankiya told TechCrunch.
Most of the significant decisions to invest in startups often came from Jensen himself, according to Umesh Padval, managing partner at Seligman Ventures. Padval has known Huang since serving on the board of Mellanox Technologies, a networking chipmaker Nvidia acquired for approximately $7 billion in 2020. While there is hope that Huang’s eventual successor will resume his dealmaking spree, investors aren’t hopeful that it will happen in the short term.
Nvidia’s interim CEO, Colette Kress, is expected to institute some changes, especially when it comes to Nvidia’s investments into AI startups at the crux of the billion-dollar data center build-out. Huang’s rampant dealmaking in this sphere has been a source of concern for Wall Street.
“She’s much more conservative,” one investor said of Kress. “I don’t think she’s going to commit the same amount of money to keep the early-stage ecosystem going.”
Nvidia’s biggest bets were backing neoclouds, inference providers, and open weight models, which are expensive for smaller companies to build. “Jensen’s initiatives are a big reason an ecosystem has emerged despite that,” Zain Javaid, co-founder of the hardware company Ghost, said.
For instance, Nvidia backed CoreWeave when it was still private and participated in two of Crusoe’s rounds, including its $1.4 billion Series E. It bought Hugging Face this year for $13 billion (expected to close in 2027) and holds investments in Mistral AI, Lambda, and Black Forest Labs. Nvidia was often the largest investor in the deals. Last year, it wrote an $800 million check into a $2 billion Series B round for Reflection AI, before the open source startup began to earn meaningful revenue, according to the Wall Street Journal.
“Jensen believed in open source and frankly speaking, Europe,” Bankiya said of how Huang invested so much in AI across the world. The cease of that support in other regions outside of America, like Europe “would be a massive funding and honestly ecosystem gap that no one can fill.”
It’s also true that Nvidia was such a dominant player in the ecosystem that at times, it felt like a monopoly. By backing so many companies with such generous checks, Nvidia effectively locked those startups into using its chips, which helped reinforce its dominant market position.
If the company does grow more conservative with its startup investments, that could leave more room for rivals and smaller competitors. (Coincidentally, Google announced two new chips this week amid Huang’s Nvidia fallout.)
Some investors with whom TechCrunch spoke hope that large cloud companies and other chipmakers will step in to back startups and fill some of the void that will likely be left by Huang.
Dharmesh Thakker, a general partner at Battery Ventures, for example, expects Big Tech rivals to vie for Nvidia’s investment crown, “depending upon who brings the best combination of a great product, an ecosystem and community to compete with CUDA, and the financial power to bankroll the ecosystem,” he said. CUDA is Nvidia’s software platform that enables developers to run AI models on its GPUs.
While companies like Google, Amazon, Microsoft, and AMD may back mature startups to drive adoption of their chips and clouds, Thakker doubts they will pour hundreds of millions into pre-revenue infrastructure startups the way Nvidia did.
Under Huang, infrastructure companies that needed $100 million to $500 million in GPUs would often raise half of that amount at valuations of $1 billion or more from Nvidia just to get started, one VC said. The chipmaker’s participation gave other investors the comfort in backing these companies at the lofty valuations.
Now Nvidia is expected to be more selective in the new companies it backs, while writing smaller-sized checks into its existing portfolio, contributing to a drop in valuations.
“When Nvidia was on the cap table, everyone felt safe,” Padval said. Without that backing, investors say traditional VCs are likely to think twice before writing checks into these startups and may even pause certain dealmaking.
Even so, few believe that Nvidia’s investment slowdown will by itself halt the broader AI boom.
Though everyone agrees that Huang is a massive loss for Nvidia, one of his ex-employees, Amulya Vishwanath, who now runs her own VC fund, said that “founders and leaders must be held accountable,” and that the ecosystem is bigger than one person. “That talent will, and must, continue to thrive.”
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.