OpenAI chief Sam Altman says his high-profile artificial intelligence company is "on the wrong side of history" when it comes to being open about how its technology works.
Altman's comments came during an Ask Me Anything session on Reddit where he fielded questions including whether he would consider publishing OpenAI research.
Altman replied he was in favor of the idea and that it is a topic of discussion inside San Francisco-based OpenAI.
"I personally think we have been on the wrong side of history here and need to figure out a different open source strategy," Altman said. "Not everyone at OpenAI shares this view, and it's also not our current highest priority."
Chinese AI newcomer DeepSeek has made headlines for its R1 chatbot's supposed low cost and high performance, but also its claim to be a public-spirited "open-source" project in contrast to closed alternatives from OpenAI and Google.
Open source refers to the practice of programmers revealing the source code of their software, rather than just the "compiled" program ready to run on a computer.
This has clashed with private companies' pursuit of revenue and intellectual property protection.
Meta, DeepSeek and France-based AI developer Mistral claim to set themselves apart by allowing developers free access to their tools' inner workings.
A member of the Reddit group asked Altman whether DeepSeek has changed his plans for future OpenAI models.
"It's a very good model," Altman said of DeepSeek. "We will produce better models, but we will maintain less of a lead than we did in previous years."
© 2025 AFP
14 Comments
Login to comment
sakurasuki
That's all happen because to DeepSeek.
LLAMA by Meta already made opensource model, so does DeepSeek.
Before DeepnSeek, OpenAI is non-profit with closed source, now OpenAI consider another position.
So the era of needs of billions development from investor, it just over.
isabelle
The key word here is "supposed." Again, we must point out that the reason DeepSeek exists is because it ripped off OpenAI.
https://www.tomshardware.com/tech-industry/artificial-intelligence/microsoft-and-open-ai-investigate-whether-deepseek-illicitly-obtained-data-from-chatgpt
Though the code is open-source, the data isn't. And that data is full of the usual CCP censorship, lies, and propaganda: claiming the Senkakus are China's, refusing to answer questions on the Uyghur Genocide, etc. etc. etc.
https://www.forbes.com/sites/maryroeloffs/2025/01/27/does-deepseek-censor-its-answers-we-asked-5-questions-on-sensitive-china-topics/
https://thediplomat.com/2025/01/why-deepseek-presents-serious-problems-for-china-researchers/
Everyone should be aware of the above (and, of course, the fact that the CCP will get your personal data) if they are thinking of using DeepSeek.
sakurasuki
Be honest how many people really use prompt for Senkaku, Uyghuer and even Tianamen?
There's many others prompt than that.
Do you think ChatGPT non-biased? Try certain prompt, you'll get surprised.
https://www.reddit.com/r/Jewish/comments/1fep4fg/chat_gpt_is_deleting_my_israel_conflict_research/
https://mashable.com/article/ai-chatbot-israel-palestine-chatgpt-google-bard
https://www.quora.com/Why-does-Chat-GPT-protect-Jews
nishikat
I just want to know the best way to bake a cake (for example). Who cares about that and I just want a cheap solution. If you need to share your most emotional secrets with American based AI or verify if Cheney remote controlled the planes in the twin towers on 9/11 then you have real issues.
Then we have a choice of CCP AI or Trump AI. Make your choice. I just want to bake a good cake. Beyond that any users have their own issues to deal with.
isabelle
On DeepSeek? Not many, as it will tell you either lies, or nothing at all.
On platforms not controlled by the CCP? Unknown, but the information is there if needed.
If you don't care about things like human rights or oppression, you can "happily" use DeepSeek and other CCP-controlled tools.
I do care, so will never use them.
False dilemma. There are a lot more choices than that.
iron man
The Big question is does any of this overcoded c...p tell us the truth anymore. All physical tabloids have only been printing want their reader's want to read for decades. Whatever it is. So alterman posts how to improve it. I see the immediate reaction. Deepseek is fake, CCP must be banished from the planet.. ban CCP. So the alterman, runaway cum back to be on of Trumps trusty Leftenants one post bang bang boom. I enjoy watching F1 cars, It gets me away from the nonsense of all this Alien intelligence. tell 'em what they want to hear!
justasking
This means AI money will be gone soon. Time to find a new job.
Pukey2
izzy:
As I said before (and as you conveniently ignored), the beauty of DeepSeek is that it is open source and allows for tweaks for local variations. So, if you want DeepSeek to come up with your usual CIA-talking points, you just need to do some tweaking for your own market, and there you are.
But then there would no reason for you to have a hissy fit or get your knickers in a twist.
By the way, has Biden's 1.6 billion dollar anti-China propaganda fund run out yet?
isabelle
I know the code (but not the data) is open source, but it doesn't change the fact that:
1) DeepSeek spouts CCP propaganda
2) DeepSeek ripped off OpenAI
And that's what everyone needs to be aware of.
Sanjinosebleed
I love it!
China leading the way here!
nishikat
Agreed and I've always been against the trade war. Peace. I'm not whining but Sam Altman is.....and that's his problem....PEACE
Time to get some good cake recipes
WeiWei
These are not facts. Especially 2).
And ChatGPT spouts other kinds of propaganda so why is that propaganda OK but CCP NOK?
isabelle
I will concede 2: incorrect phrasing. I should have said "seemingly ripped off" and quantified that with "but the evidence is already pretty damning," as I did on another thread.
But I certainly won't concede 1, as the output is self-evident. Anyone not blinded by CCP ideology can see that, and it's ludicrous to argue otherwise.
https://thediplomat.com/2025/01/why-deepseek-presents-serious-problems-for-china-researchers/
Propaganda isn't OK. But if anything is incorrect on ChatGPT, it can be flagged and corrected.
DeepSeek will never have its propaganda corrected, as it must adhere to the CCP line.
quercetum
Silicon Valley 矽谷 is now well behind Shenzhen 深圳 and Guangzhou 廣州。There is also Beijing 北京 and Hangzhou 杭州 to contend with. China will commercially mass produce "advanced chips" at a fraction of NVDIAs production capacity. 2025 to 2030 will be exciting.
DeepSeek’s story isn’t just about AI. It’s a case study in how China turns adversity into opportunity. They are a start up and do not have billions of dollars. Think about it. Whenever the US sanctions or bans China e.g. space stations to solar farms, it leads to homegrown breakthroughs.