OpenAI has officially entered the custom silicon race, unveiling its first AI chip, dubbed Jalapeño, in a move that analysts say threatens Nvidia’s near-monopoly on advanced AI semiconductors.
The chip, designed specifically for inference — the process of running AI models day-to-day — was announced Tuesday with claims of “industry-leading speed and efficiency,” according to CNBC. It will be deployed within OpenAI’s own compute infrastructure by the end of the year, with second and third generations already in development alongside Broadcom.
Inference efficiency on par with Blackwell
Adrien Sanchez, technology analyst at Yole Group, told CNBC that Jalapeño demonstrates a “hyperscaler-designed chip can now match or beat Nvidia’s Blackwell-class GPUs on inference efficiency.” While Nvidia still commands the “vast majority” of AI compute and benefits from deep ecosystem lock-in through its CUDA software platform, Sanchez called the new chip a “threat to Nvidia’s inference margins, which is the field growing the most at the moment.”
Research firm SemiAnalysis said it visited OpenAI’s labs to benchmark the chip, finding it beat Blackwell on performance per watt in nearly all tested scenarios. However, the firm cautioned that the comparison was “somewhat incomplete and unfair” — Jalapeño uses newer HBM4 memory, while Nvidia’s Rubin platform, which also uses HBM4, is a more like-for-like comparison.
“Jalapeño is really competing against chips like Rubin that also use HBM4,” SemiAnalysis analysts wrote in a blog post on Tuesday, noting that “Vera Rubin systems are starting to ship to customers right now, while it will still be some time before OpenAI has anything beyond engineering samples of Jalapeño.”
Efficiency gains could reshape unit economics
Alexander Harrowell, senior principal analyst at Omdia, described Jalapeño as an “impressive achievement, most of all in terms of efficiency.”
“In a large-scale deployment, this would save power, cooling, and power distribution infrastructure, and contribute a lot to their unit economics,” he told CNBC.
OpenAI said the chip would allow users to get “faster responses, more responsive agents and more reliable access” as demand for its AI services grows.
A broader shift toward custom silicon
OpenAI is far from alone in building its own chips. Google has unveiled its tensor processing units (TPUs) for AI training and inference, Meta has agreed to deploy 1 gigawatt of custom AI chips using Broadcom technology, and Anthropic committed to spending more than $100 billion on AWS tech over the next decade, including Amazon’s custom Trainium chips. Startups such as Cerebras, SambaNova, D-Matrix, Etched and Fractile are also developing AI-specific processors.
Harrowell noted that Omdia expects custom ASIC chips like Jalapeño to exceed GPUs in volume by 2028, though GPU revenue will remain larger for some time given their considerably higher price tags.
“This is the biggest competitive threat to Nvidia, as about half the capital expenditure on AI infrastructure comes from hyperscale cloud providers who either have a custom chip program or could reasonably have one,” he said.
Implications for Nvidia’s largest customer
OpenAI has historically been one of the largest buyers of Nvidia GPUs, using them to train and run massive AI models. Having its own inference chip could reduce that reliance over time, according to TrendForce analyst Fion Chiu.
But Chiu added that for compute-intensive workloads like large-scale model training and frontier AI, “we believe Nvidia GPUs will remain important given their broad programmability, performance, software ecosystem, and ability to handle a wide range of workloads.”
Sanchez framed the development in stark terms, saying Jalapeño “raises the stakes for Nvidia’s largest customer relationship specifically.”
Source: www.cnbc.com — https://www.cnbc.com/2026/08/26/openai-jalapeno-ai-chip-nvidia.html
This article is for informational purposes only and does not constitute financial, investment, tax, or legal advice. Do your own research and consult a licensed professional before making financial decisions.
