Reality is, these AI giants are here and they are using a massive amount of resources. Love them or hate them, that is where we are. Whether or not you accept the job with them, OpenAI is gonna OpenAI.
Given how much the detractors scream about resource uses, you'd think they'd welcome the fact that someone of your calibre is going in and attempting to make a difference.
Which, leads me to believe you're encountering a lot of projecting from people who perhaps can't land the highest of comp roles, and shield their ego by ascribing to the concept of it being selling out, which they would of course never do.
However. I am putting my curious foot forward here:
What were the toughest ethical quandaries you faced when deciding to join OpenAI?
Comments like this one and at least two of your recent blog posts makes you look like a grifter.
I loved your work back when I was an IC, and I'm sure this is a common sentiment across the industry amongst those of us who started systems adjacent! I still refer to your BPF tools and Systems Performance books despite having not written professional code for years now.
Can't wait to read content similar to what you wrote about when at Netflix and Intel albeit about the newer generation of GPUs and ASICs and the newer generation of performance problems!
Nobody is hating anyone here, I don’t know where you got that from.
They are asking Brendan to be honest.
It is fine to say it is about the money.
But it is silly, borderline patronising to tell readers that you’re “saving the world” because you got a job at OpenAI.
I hope there will be harder problem waiting for you, than using flamegraphs to optimize GenAI Porn.
https://www.axios.com/2025/10/14/openai-chatgpt-erotica-ment...
What do you think is happening with the efficiency gains? You're making rich people richer and helping AI to become an integral (i.e. positive ROI from business perspective) part of our lives. And that's perfectly fine if it aligns with your philosophy. It's not for quite a few others, and you not owning up to it leads to all kinds of negativity in the comments.
may it happen that the efficiency gains decrease demand and thus postpone investment into and development of new and better energy sources? If one couldn't get by just by bringing 20 trucks with gas turbines, may be he would have invested in fusion development :)
What mechanism would make this happen?
Demand could decrease if AI became worse, but efficiency doesn't make AI worse - it actually makes possible at all to run bigger, better models (see the other comment with a link to Jevon's paradox), which increase, not decrease demand (more powerful models may have new capabilities that people want to use)
Alternatively, AI demand could decrease through political pressure (either anti-AI sentiment takes a foothold on the public, and/or government regulation strangle demand on the sector like it did for eg. on tobacco industry). But another way to reap the benefits of more efficient AI datacenters is to make it a talking point on how AI environmental impacts can be mitigated, which could curb anti-AI sentiment.
Either way, those possibilities don't decrease demand for AI - they are either neutral, or increase demand instead.
You have done a brilliant job elevating your chosen specialty to the world, and encouraging and inspiring others in the industry for a long time - so you should be fairly compensated for that lofty position. I don't envy the late nights or very early mornings you have ahead of you on conference calls with SF, but good luck at OpenAI mate !
Of course once someone has money they can say it's not about the money, but that privilege is literally bought with...money.
Reducing runtime energy use over years won't really make up for the resource use that goes into building the data center. It's just moved around, similar to how Elon moves around money as needed to bolster the financials of a particular project.
Like with the airline industry it's not just the smog they blow on our food. Drink carts, seat belts, barf bags all have a resource intensive energy and materials pipeline.
Every server screw and power cable adds up.
Quite frankly, I think some people here are too quickly spooked and think what you say is sus. I simply see that as a sign that they aren't fully having a good faith discussion. Or they simply read things way differently than I do.
I'm simply writing this because I think there are enough people that have a similar reading to what you wrote. They simply don't mention it as people who feel "outraged" (a bit too dramatic of a term but English is my 2nd language). "Outraged" people seem simply more vocal to me.
For clarity: I feel neutral about this whole thing. I do appreciate the work you've done in the past.
I'm surprised at this; all that experience wasn't enough to flag this article as obviously AI generated?
More to the point, with all that experience you still weren't able to issue prompts to make the output sound different from generic AI slop?
I am a long time fan, I have the physical copy of each and every book that you have authored, I have watched each and every video that you are in, and I walk team members and clients through your USE method at every engagement I am on.
I would say to you that the "make the world a better place" has been excessively misquoted. Even the Silicon Valley episode on Tech Crunch parodies show how anything and everything is intended to "make the world a better place".
Please reconsider your use of the phrase given the well-earned negativity around it.
I actually said "it's about saving the planet" based on all the environmentally worried news articles I've seen.
It does appear that for many of us "saving the planet" has become akin to "making the world a better place".
I was excited last night thinking about what you might uncover at OpenAPI. While we can all speculate (device drivers? File systems? Python itself? DB queries? Memory architecture? Inference algos? ) I recall “I have no data yet. It is a capital mistake to theorize before one has data” by Sherlock Holmes in "A Scandal in Bohemia".
All the best with your work at OpenAPI. We will all hopefully learn about writing more energy efficient code and about more efficient LLMs and more thanks to your work!
I’ll give it a shot; I think you’re successful in what you do and very altruistic and open, not only in your discoveries but also your opinions. You also have a higher sense of duty. as oscar wilde once said, we’re all in the gutter, but some of us are looking at the stars. compensation is boring gutter talk. It’s hard for people to reconcile your benevolence with your success, and just as the trope of joining a company to change the world is a veil for making money, so is the trope of criticizing the agent of change entering an industry because the industry is bad.
Personally I can’t wait to read about the inefficiencies you find and have a little glimpse into openai tech from your opinionated point of view.
What a fucking narcissist.
> There's so many interesting things to work on, things I have done before and things I haven't.
What are the things you haven’t done before, if you could mention them?
Brendan.
First of all congratulations on your new job. However,
It is easier to just say to everyone it is about the money, compensation and the stock options.
You're not joining a charity, or to save the planet, this company is about to unload on the public markets at an unfathomable $1TN valuation.
Don't insult your readers.
EDIT: possibly a corollary--does Mia pay money for chatgpt or use a free plan?
My wife was paying for ChatGPT before I joined. I didn't ask Mia. I probably have three months of hair growth before my next chance to ask.
I find it a little amusing that AI companies provide free AI subscriptions to their employees and their families. Perhaps because I'd never thought of it that way.
That's kinda nice actually; I assume employees get early access to models to test (5.3 codex, for example). Do families get it too?
The string "compens" appears exactly once in your post:
> But there are other factors to consider beyond a well-known product: what's my role, who am I doing it with, and what is the COMPENSation?
You did it for the money; don't try to rationalize it, because no one believes you. For that amount of cash, I'd probably jump on Altman's bubble for a year or two.
Wanna buy a bridge?
No, it never does. Those people somehow delude themselves into thinking it might, but...it might just work for us.
This is a company which at the first opportunity seized and stopped doing open research, cut open source contributions, converted itself to for profit after years of fiscal benefits, that scrapped its ethics committee and removed all engineers who opposed any of this.
Don't come with the excuse there is any work being done for the better of something.
One should never input one's own expectations into another, but, I feel disappointed. It is having the guy I saw growing from the first posts working for an evil machine on his own volition.
Do what you want. But that's what I feel about this disheartening news
Inferring the overall tone from the comments, I think the folks here are struggling with what sounds like a logical fallacy from someone who is certainly a logical thinker.
> how I could lead performance efforts and help save the planet.
The problem on the face of it being: Performance gains will not translate to less energy usage (and by extension less heat released into the atmosphere). Rather, performance gains will mean that more effective compute can be squeezed from the existing hardware.
If performance gains translate to better utilization of the hardware, it also follows that it will translate to more money for the company, allowing for the purchase of more GPUs. Ad infinitum.
My stance is that this is just businesses doing what they do. It's always required regulation to slow down the direct/indirect negative byproducts (petro companies being the most obvious example). I don't see how AI would inherently be different.
Is there another angle that I (we) am (are) missing where the performance efficiencies translate to net benefits for the planet?
As a performance engineer I'm familiar with Jevons paradox, but it does not discourage improving efficiency.
Would be fantastic if you can find a way to make optimizations you find available more openly. The whole ecosystem benefits when efficiency improvements are shared. Looking forward to seeing where this goes and don't let the negativity from some get to you
You arnt going to stop the excesses.