Strategic Partnership Aims to Speed Up Computing and AI Engineering
Nvidia made a significant move on Monday, acquiring a $2 billion stake in Synopsys to deepen their partnership and accelerate computing and AI engineering. The shares were purchased at $414.79 each, according to company figures.
This collaboration is designed to shift intensive design work from slower systems to GPU-powered workflows. The deal was announced during a dynamic week in the AI sector, with continued strong investment in technologies related to computing speed.
The market responded quickly to the news. Synopsys shares saw a 4% increase on the day, while Nvidia's stock rose by 1%. The timing of the announcement is notable, as Synopsys shares have climbed 33% this year but have also experienced a nearly 12% decline this month, highlighting the volatility and rapid pace of the AI landscape.
Nvidia is a key provider of tools for training large AI systems, while Synopsys offers the software used to design the chips that power these systems. This partnership bridges these critical aspects of the AI ecosystem.
Expanding GPU-Driven Design Workloads
The multi-year partnership will focus on compute-intensive applications, agentic AI tools, cloud accessibility, and joint market initiatives. Nvidia will contribute its hardware expertise, while Synopsys will integrate more of its design software onto accelerated systems.
The primary objective is to enhance the speed of handling massive workloads, rather than altering the existing sales structures of either company.
Nvidia CEO Jensen Huang commented on the deal during an interview on CNBC, emphasizing its focus on the design and engineering sector, which he described as one of the most compute-demanding industries globally. He noted a significant industry shift from traditional CPU-based systems towards GPU-accelerated computing.
Huang clarified that while CPU systems will continue to be relevant, the majority of heavy computational tasks are migrating to accelerated platforms.
Synopsys CEO Sassine Ghazi highlighted the transformative impact of this partnership, stating that tasks previously requiring weeks can now be completed in mere hours. This acceleration is crucial for chip layout, silicon verification, power modeling, and system routing—processes that often delay hardware launches and increase costs.
The relationship between Nvidia and Synopsys has a long history, with Huang mentioning that Nvidia itself was built using Synopsys design tools. This new agreement maintains that historical connection and remains non-exclusive, allowing both companies to continue collaborating with other industry partners.
Nvidia's continued success in the AI build-out stems from its GPU sales, essential for training AI models and managing large workloads. Synopsys complements this by providing silicon design and electronic design automation software, creating a comprehensive solution from chip conceptualization to deployed AI systems.
Wall Street Analysts Raise Concerns Amidst Rising Competition and Spending
Despite widespread optimism on Wall Street, Seaport maintains a sell rating on Nvidia, with analyst Jay Goldberg reiterating this stance in a recent note. Goldberg's price target of $140 is approximately 21% below Nvidia's previous closing price of $177.
Goldberg acknowledged the strength of Nvidia's business but pointed to the AI boom creating complex sales structures and opaque accounting practices. He specifically cited $26 billion in prepaid cloud compute costs on Nvidia's balance sheet, which the company attributes to research and cloud services for its DGX platform.
Goldberg questioned whether research alone would justify this amount, suggesting it might represent rebates tied to large customer deals. These deals could include provisions where Nvidia agrees to purchase excess capacity from its clients if needed.
Nvidia's working capital has also increased significantly, which the company interprets as a sign of strong demand. However, Goldberg views this as a double-edged indicator, especially when combined with rising customer commitments.
He noted that Nvidia has invested $6 billion in private companies this year and holds an additional $17 billion in commitments, including $5 billion allocated to Intel. Furthermore, an unsigned agreement with OpenAI could potentially add another $100 billion if finalized.
Goldberg suggested that Nvidia might recoup these funds when these companies secure capital and purchase more systems. Nevertheless, he warned that the sheer scale of these investments indicates increasing pressure from competing chip manufacturers.
He also highlighted the growing competition from Google's in-house TPUs. Goldberg stated that these systems have demonstrated superior performance to Nvidia's hardware in certain benchmarks, although they may not be suitable for all clients. He also observed that Google is actively promoting TPU adoption among external partners.
According to Tipranks data, out of 66 analysts covering Nvidia, 59 recommend a buy or strong buy, six suggest holding, and only one, Seaport, rates the stock as underperform.