Sachin Katti left Intel in November 2025 to join OpenAI, where he is working on the compute infrastructure needed for OpenAI’s research and products. The move came only about six to seven months after Intel CEO Lip-Bu Tan appointed Katti chief technology and AI officer. Intel said Tan would take oversight of its AI and Advanced Technologies Groups after Katti’s departure.
The transition is significant because it moves a senior leader with experience in networking, silicon strategy, edge computing, and AI from a chipmaker to one of the world’s largest prospective buyers and builders of AI infrastructure. It does not, however, prove that OpenAI hired Katti to design a particular processor or that Intel’s AI strategy has failed.
As an Amazon Associate I earn from qualifying purchases.
What happened to Sachin Katti?
Intel confirmed Katti’s departure in November 2025, while Katti said he was joining OpenAI to help “build out the compute infrastructure for AGI.” Greg Brockman described the work more broadly as helping design and build the infrastructure required for OpenAI’s research and applications.
Recommended Free Tools
Bloomberg later described Katti as OpenAI’s head of industrial compute. Some secondary coverage has used “Head of Compute Infrastructure,” but OpenAI has not publicly established that exact title in the cited material. The safest description is that Katti is an OpenAI infrastructure executive working on compute and industrial-scale capacity.
#1 Best Overall
Reports published on November 10 and 11, 2025 establish the month of the move. They do not establish an exact final employment date or a definitive personal reason for his departure.
CRN reported on Intel’s confirmation and Katti’s move, while Data Center Dynamics quoted Katti’s description of his new remit.
Katti’s role at Intel
Katti joined Intel around 2021 and became general manager of its Network and Edge Group in early 2023. In April 2025, Tan elevated him to chief technology and AI officer while he continued leading Network and Edge.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →His responsibilities covered more than corporate AI messaging. Intel’s Network and Edge work included 5G systems, virtualized radio-access networks, Xeon-based edge platforms, networking, and AI acceleration. Katti’s background also connected him to broader product and silicon-roadmap decisions.
Rank #2
In June 2025, Intel announced additional AI-silicon and networking leadership reporting to Katti. That structure indicated that his remit included elements of AI product strategy, AI system-on-chip engineering, networking, and infrastructure architecture.
Intel’s leadership announcement supports Katti’s title and the reporting structure. Intel’s Network and Edge material describes the areas associated with his organization.
What will Katti do at OpenAI?
The confirmed remit is compute infrastructure for AGI research and the scaling of OpenAI’s applications. In practical terms, that kind of responsibility can span:
- data-center capacity and power availability;
- accelerator and chip supply;
- high-speed networking and interconnects;
- memory and storage systems;
- cloud partnerships and custom hardware;
- systems software, reliability, and operational scaling; and
- matching infrastructure to model-training and product workloads.
That list describes the demands of the job, not a published OpenAI project plan attributed specifically to Katti. There is no cited evidence that he was hired to design a proprietary OpenAI processor, replace Nvidia, select Intel hardware, or report directly to a particular OpenAI executive.
OpenAI uses “AGI” to describe a general-purpose level of artificial intelligence. The term is part of OpenAI’s stated mission; it should not be read as evidence that AGI has already been achieved or as a settled scientific category.
Why OpenAI needs infrastructure leadership
Large AI systems are constrained by more than the number of available GPUs. A useful cluster must also have sufficient power, cooling, networking, memory, storage, software, reliability, and physical data-center capacity. A bottleneck in any one of those areas can reduce the value of the accelerators already installed.
Katti’s experience is relevant because networking and edge systems require coordinating hardware, software, communications, and workload performance. That background does not prove that OpenAI has selected a particular architecture, but it helps explain why an AI company would recruit from a semiconductor and infrastructure organization rather than only from a model-research team.
Free tools Windows power users keep installed
One-click scans. No signup required.
OpenAI’s public infrastructure announcements show the scale of the challenge. OpenAI said in an April 2026 update that its Stargate effort had committed to securing 10 gigawatts of U.S. AI infrastructure by 2029 and had surpassed that milestone, according to OpenAI’s own account.
OpenAI also announced a $38 billion AWS infrastructure agreement in November 2025 and a Broadcom collaboration involving 10 gigawatts of OpenAI-designed accelerators in October 2025. Those announcements demonstrate the breadth of OpenAI’s infrastructure strategy. They do not show that Katti personally negotiated or led either arrangement.
What Intel loses
Intel loses a leader who had recently been given a broad AI and technology mandate at a time when the company was trying to strengthen its position in data-center AI hardware and software.
Intel’s immediate response was for Tan to assume oversight of the AI and Advanced Technologies Groups. That suggests Katti’s responsibilities were substantial, but it does not establish that Intel’s product roadmap has been delayed, canceled, or materially damaged.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11The clearest risk is continuity. Katti had been associated with AI strategy, networking and edge operations, silicon-roadmap coordination, and newly appointed AI-systems leaders. Intel must now maintain alignment across those areas while competing with Nvidia, AMD, hyperscalers, and custom-chip programs.
Best Value
One executive’s departure is therefore best understood as a leadership and execution challenge—not conclusive evidence that Intel is losing the AI race.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Why Katti was a logical hire
Before Intel, Katti was a Stanford electrical-engineering and computer-science professor and held a doctorate from MIT. He is also an IIT Bombay alumnus and co-founded wireless-technology companies including Kumu Networks and Uhana.
That combination gives him experience across academic research, wireless communications, networking, startups, edge computing, and corporate silicon strategy. Those skills are increasingly valuable as AI companies build infrastructure that must operate as an integrated system rather than as a collection of separately purchased chips.
The move also illustrates a broader shift in the AI industry: model developers are hiring leaders from chipmakers, cloud providers, networking companies, and data-center operators because access to compute has become a strategic constraint.
What remains unknown
- The exact OpenAI title: Bloomberg’s “head of industrial compute” description should be attributed. “Head of Compute Infrastructure” should not be treated as an officially confirmed title without further confirmation.
- The reason for the move: The public record does not establish whether compensation, recruitment, strategic disagreements, or dissatisfaction played any role.
- The reporting structure: There is no cited confirmation of whom Katti reports to.
- OpenAI’s hardware plans: The company’s Broadcom collaboration concerns OpenAI-designed accelerators, but it does not establish that Katti personally designed them or that OpenAI is abandoning Nvidia, Intel, or other suppliers.
- Intel’s future results: The departure alone cannot show whether Intel’s AI products, sales, or roadmap will suffer.
Bottom line
Sachin Katti’s move from Intel to OpenAI is a meaningful signal about the importance of AI infrastructure. OpenAI is assembling the compute, power, networking, cloud, and custom-hardware capacity needed to support its ambitions, while Intel is trying to coordinate its own AI-silicon and data-center strategy.
The evidence supports describing Katti as an OpenAI infrastructure leader working on industrial-scale compute. It does not support claims that he was hired to build a specific chip, that Intel’s AI effort has collapsed, or that his personal move determines which hardware OpenAI will use.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
