Intel’s Open-Source Strategy Is Changing At Odds With The Ethos Of Open-Source

For the past 21+ years of running Phoronix and even longer than that being a Linux user, I have loved and consistently promoted Intel’s open-source efforts and leading Linux support. Even through Intel’s difficult periods of delayed and stagnate hardware launches, what had remained consistent at the company and rather legendary had been their open-source contributions. From the Linux kernel to compiler toolchains and hundreds — if not thousands — of different open-source projects over the past two decades have been advanced thanks to Intel’s open-source leadership. It is with much sadness that my faith and confidence in Intel’s open-source leadership position is being questioned and questioning the direction they are now apparently steering their open-source focus/philosophy moving forward.

I have passionately enjoyed Intel’s contributions to the open-source ecosystem and eagerly promoted them over the years. At technical events I’d be eager to chat with Intel engineers to hear about their latest open-source work whether it be working on new Linux kernel optimizations, upstreaming of upcoming hardware support, or contributions to XYZ open-source project for ensuring leading performance on their upcoming wares. Talking software optimizations and other open-source work is consistently exciting for dedicated engineers compared to the annual or bi-annual hardware launch cadences. Prior to recent restructuring at the company, it seems like it would be every few days hearing about some interesting open-source software contribution worthy of writing about on Phoronix with much personal intrigue and often the Phoronix content leading to broader excitement in the open-source ecosystem. There have been literally thousands of articles on Phoronix over the years about Intel’s stellar open-source efforts. Intel’s open-source leadership over the past two decades is rather undisputed: AMD has since ramped up their open-source efforts too with the continuing successes of Zen and their other products, but in the pre-Zen era there was absolutely no comparison that Intel did more than their lead rival around Linux/open-source. Intel also has done a lot more open-source work than the likes of Arm, RISC-V vendors, and other CPU ISAs/organizations with just the possible exception of IBM. Intel’s incredible open-source work had led many to continuing to purchase Intel hardware even during the long-lived quad-core 14nm era and other hardware setbacks at the company since their Linux hardware support was consistently great and Intel was known for being so prolific in the open-source world. Intel’s open-source software engineers have arguably been the company’s greatest sales tool for Linux users/organizations.

Intel’s open-source software strategy though has been clearly changing given the corporate struggles and subsequent restructuring at the company. Intel has been losing significant numbers of their Linux/open-source software engineers over the past year as has been covered in many articles…. During the continuing AI boom, the Habana Labs AI driver was on hiatus the past year after losing several rounds of driver maintainers. Many Intel Linux kernel maintainers were laid-off or otherwise departed. This year Intel lost some of their brightest open-source engineers to the likes of Meta and NVIDIA. Even the Intel CPU temperature monitoring driver for Linux was marked “unmaintained” following layoffs, among other Intel Linux drivers being orphaned. Outside of kernel space, some user-space projects have been sunset. Intel’s reduction in open-source/Linux staff has also led to many Debian packages for Intel software/hardware being orphaned with no apparent Debian DDs/DMs left on the Intel team. A heavy-hitting change was also Intel ending Clear Linux for what consistently delivered the best Linux x86_64 performance over the past decade and pains me still they sunset such a forward-looking project. As Intel is able to regain hardware competitiveness, their abandoning of Clear Linux and other software shifts will likely prove regret.

The Open-Source Statements At Intel Tech Tour Arizona

Beyond shedding developers, it is evident now as a result of statements made by Intel leadership, their open-source doctrine is shifting. To much frustration given my passion for Intel open-source over the years, last week at the Intel Tech Tour in Arizona statements were made that Intel will look at limiting/refraining from open-source contributions if in effect it means potentially helping their competitors or rather how to position open-source for Intel more favorably from their competition. Kevork Kechichian who is the new leader of Intel’s EVP and GM Data Center group spoke during his keynote pertaining to the upcoming Xeon 6+ Clearwater Forest announcement that the company needs to figure out how to make open-source more of an advantage to Intel and not its competitors. These words were bad enough to hear from any Intel executive given the company’s vast open-source history, and not from some Windows-focused client representative but from a data center executive. In 2025 where open-source software is so pervasive throughout the ecosystem, especially in the era of AI and the hyperscalers…

Kevork stated when talking about providing customers with end-to-end solutions, “an era where platforms are super important”, and on Intel investing in open-source infrastructure:

“We need to find a balance where we use [our open-source software] as an advantage to Intel and not let everyone else take it and run with it.”

At first questioning to myself what I had just heard… He further added a few minutes later in his presentation, Kevork stated as part of making a successful data center business at Intel:

“We are very proud of our open-source contributions. We are going to keep on doing that. However, like I mentioned, I want to make sure that it gives us an edge against everyone else.”

At the very least I don’t even understand why these statements were made. Open-source software didn’t even need to get brought up in the context of announcing Xeon 6+ / Clearwater Forest. All that could have been said is that Intel is going to pursue hardware excellence for regaining its dominant position, but instead making sure open-source software contributions aren’t being better leveraged by competitors than Intel was cited. Twice. This could have not even been brought up, as what customers are going to get excited about a less open-source friendly Intel?

Intel’s open-source contributions have historically not been limited to making sure it gives them a unique advantage but benefiting the Linux landscape as a whole or at least the x86_64 world, ala a rising tide lifts all boats. Intel has done so much to help the Linux kernel in all areas from device drivers to memory management and other low-level plumbing. Thus now an apparent change in direction of open-source at Intel from anything else I’ve heard from any Intel executive before.

What Will Be The Impact?

These statements were baffling to me. Intel expects to regain a competitive edge and help become more competitive in the industry by — in part — limiting their open-source contributions if it could help competitors or not give Intel an advantage? Considering the x86_64 ISA, this could mean a pretty big impact. Will Intel limit their excellent contributions to the GCC and LLVM compiler toolchains since AMD stands to benefit from their x86_64 ISA additions and new compiler optimizations? Will they now gate new ISA features/optimizations behind their DPC++ compiler instead? What about all of their generic Linux kernel contributions they have done to memory management, security, and many common low-level areas of the kernel since often times they benefit other hardware too beyond just Intel? What about all of their higher-level contributions to different open-source applications and frameworks because they also benefit AMD as well as all computer users at large? What about general power management optimizations to the Linux kernel or related components that could help their competitors too with better power efficiency? (Or does that go against their separate environmental goals?) What about contributing to common Mesa code that benefits other GPU hardware vendors thanks to all the common/shared Mesa / Gallium3D / Vulkan run-time code but not giving Intel a unique advantage? What about new AVX-512 optimizations to software that helps with better compute performance and in turn also better power efficiency for the masses without any distinct advantage for Intel? (Let’s not forget about the x86 Ecosystem Advisory Group… How will Intel’s open-source contributions play out there or not, since after all AMD is a “competitor”. But Intel’s real focus should be on not ceding marketshare to ARM and RISC-V that would be harder for them to ultimately regain once organizations depart from the x86 world.) Will the new Intel-primacy open-source focus lead to a new level of de facto vendor lock-in moving forward? I have many more questions than answers at this point around what they will restrict on open-source contributions moving forward or where they will be diverting their software efforts.

With the pervasiveness of open-source software and Intel’s prior investments across the entire open-source ecosystem, there are so many areas where Intel could reduce or stop their open-source contributions as it could ultimately help their competitors or not provide them a unique and competitive advantage in the marketplace. This goes against the open-source ethos of collaboration, transparency, and shared development. I was very taken aback that Intel in 2025 would now consider restricting open-source and positioning whether open-sourcing something or not as a means of battling increased competition. I was very surprised with what I heard and followed up with others in attendance. In additional talks that some had with Kevork, it was further alluded to that moving forward Intel may refrain from making open-source contributions if it could help their competitors.

From an Intel software engineer who left the company some months ago, they had written to me at the time with similar concerns/statements foreshadowing this announcement. At the time I just assumed it to be isolated to the one individual Intel software project they were involved with… But now given these executive statements, it seemingly is a shift in company-wide policy/thinking. The engineer mentioned that one of their greatest challenges was to justify software development costs when the top performing silicon belongs to another company and not Intel. Coincidentally or not, that particular Intel software project now hasn’t seen a new release in more than one year nor any new code commits. Or another separate example is Intel taking the open-source Hyperscan software and making it proprietary licensed software.

Intel’s open-source philosophy of the past was that they weren’t doing anything to help or hurt their competitors. Under the open-source leadership during the likes of Dirk Hohndel and Imad Sousou, Intel’s open-source contributions were never limited to making sure they wouldn’t have the potential of helping competitors. Dirk and Imad were great open-source leaders while taking into account corporate environments at Intel. Even Pat Gelsinger previously called out the importance of open-source and didn’t caveat it around competitors. Others at Intel over the years like Raja Koduri were also strong proponents of Intel in the open-source world without compromise and leaving no transistor left behind with the importance of software. Arun Gupta was another open-source champion at Intel that only while working on this article did I find out he’s no longer at Intel since July and formerly was their VP and GM for Open-Source Ecosystem.

^^ While working on this article I was also going through some of my old slides/content and prior Intel open-source coverage around their acknowledging the importance of open-source everywhere and making their software pervasive. Even back in 2007 they acknowledged this with Threading Building Blocks when they open-sourced it and then taking it beyond just Intel processors. There have been so many times over the years where their engineers prided on delivering the best software everywhere without a focus on just delivering an exclusive advantage for Intel. Isn’t that part of the open-source ethos after all? James Reinders who led the TBB efforts was yet another one of their very talented engineers they lost in the past year.

Intel’s Clear Linux project ran great on AMD x86_64 hardware, countless compiler and kernel optimizations helped both Intel and those from other hardware vendors, Intel’s prior SVT-AV1 / SVT-HEVC / SVT-VP9 projects led to much faster CPU-based video encoding everywhere, the Cloud Hypervisor project started by Intel years ago is now big on ARM CPUs too, the OpenVINO AI toolkit works great on AMD and ARM, Sound Open Firmware as another project started by Intel has since been adopted by several other SoC vendors, various oneAPI software projects work well on AMD / Arm / NVIDIA / IBM hardware, and the list goes on and on of how Intel previously raised the open-source software ecosystem as a whole. Intel’s shortcoming wasn’t that they invested too much in open-source or helped their competitor software stacks too much but that they failed to deliver the best processors often times for different workloads and customer needs or were too late to the party. Open-source is not Intel’s problem but arguably just opened more doors for the company over the years thanks to their monumental open-source work from creating passionate customers to helping the larger organizations deploy Intel-optimized software.

Intel also used to host a wonderful open-source conference during their years of the Open-Source Technology Center, 01.org, etc to help foster the community. Plus they have sponsored countless open-source projects and events over the years. Outside of Intel’s open-source work on hardware-specific device drivers and enabling use of new CPU ISA extensions not found elsewhere (i.e. no AMD support, yet), the scope of open-source work by Intel that wouldn’t potentially help “competitors” is rather small. After all, Intel is in the hardware business and their focus should be on shipping and selling the best hardware possible rather than artificially limiting their open-source contributions on the basis of potentially helping competitors or not giving Intel a unique advantage.

I have no doubt Intel will continue to provide timely support for new hardware on Linux, but now I have concern around their open-source leadership eroding if they will now be imposing limitations on what they open-source and depart from their previously legendary status in the open-source world for lifting all boats rather than focusing solely on their own. There is no doubt a hit already from simply the engineering talent they have lost from the corporate layoffs of the past number of months and other engineers voluntarily departing, but a shift in strategy away from their open-source ideals is shortsighted and eroding the great open-source trust they had established over the years. Just look at how many Linux/open-source users will vote with their wallet and buy Intel CPUs during the stagnate times all because of Intel’s historical open-source actions. The statements I heard last week were unfortunate and painful and will potentially damage the open-source goodwill Intel established in the community over the past many years. If there was one topic beyond Linux and open-source itself that have been consistent on Phoronix over the past two plus decades, it’s been about Intel’s open-source contributions.

I am saddened by this apparent Intel shift as it pertains to open-source software. At the very least there will simply be less open-source contributions from Intel given their reduced engineering resources but at worst a former open-source prince now more hostile toward open-source if it means potentially helping a competitor. This goes against the conventional open-source ethos and Intel’s open-source practices over the past many years of greatly enriching the open-source ecosystem.

Another Statement From Intel

I talked with several Intel representatives at the Arizona event in being shocked by these comments. Eight days later and less than 24 hours prior to the embargo lift, I received a follow-up statement from Intel on the matter:

“Intel remains deeply committed to open source. We’re sharpening our focus on where and how we contribute—ensuring our efforts not only reinforce the communities we’ve supported for decades but also highlight the unique strengths of Intel. Open source is a strategic focus designed to deliver greater value to our customers, partners, and the broader ecosystem.”

Intel can remain committed to open-source software for their own (product) needs and as mentioned I have no doubt they will continue with their open-source Linux device drivers for their own hardware (though there have been some exceptions, such as Intel IPU 7.5 With Panther Lake Will Rely On Closed-Source Libraries). But given the layoffs/restructuring in their software groups and Kevork’s comments, the new statement unfortunately doesn’t do too much to satisfy the concerns. In fact, the statement acknowledges changes are happening (“we’re sharpening our focus“) and points to more of a focus aligning with Kevork’s comments with “highlight the unique strengths of Intel.” Their open-source software has always shown the strengths of Intel from AVX-512, AMX, etc… Just less compelling when their hardware story hasn’t matched as well as their open-source ambitions. Clear Linux was a wonderful example as a means of showcasing Intel’s unique strengths and performance prowess, for which they ceased development.

Time will tell and here’s hoping for the best — both for Intel and the open-source world they helped craft so well over the past two plus decades.