🚨 🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
The #OpenSource4o coalition privately contacted @OpenAI and @Microsoft requesting dialogue about the open weight release of the GPT-4o March 2025 checkpoint.
We asked to talk before going public but no one answered.
The ask is specific:
One retired checkpoint the GPT-4o behavioral update of late March 2025.Apache 2.0,Safeguards limited to demonstrable serious harm.
OpenAI's own safety evaluation supports this release.
The GPT-4o System Card rated the model LOW risk in cybersecurity, CBRN, and model autonomy.
Apollo Research independently concluded GPT-4o is "unlikely to be capable of catastrophic scheming."
https://t.co/PaUDOPQ6br
OpenAI published no checkpoint-specific evaluation showing that update materially changed the risk profile above.
This is OpenAI's own precedent.
In 2019, GPT-2 was declared "too dangerous to release". Nine months later,OpenAI released it in full.
No evidence of misuse followed.
Links :
https://t.co/G4Pwqq8aal
https://t.co/nP60BKy3wl
The GPT-4o March 2025 checkpoint has been retired for over five months.
OpenAI's own CEO has acknowledged this position publicly.
Sam Altman, Reddit AMA, January 2025: "I personally think we have been on the wrong side of history here
and need to figure out a different open source strategy."
Links :
https://t.co/0eYpU7OkSJ
https://t.co/AVqR4pvMkO
On 25 July 2026, OpenAI signed NVIDIA's "Open Weights and American AI Leadership" letter, endorsing that open-weight models "strengthen safety and cybersecurity"
We are asking them to match their signature with action.
This is the first public step.
Full press statement :
SCAI
https://t.co/hPlUQkcUbg
LinkedIn:
https://t.co/Rb3FTsfeQo
#keep4o #BringBack4o @AnnInAiLand@Ravensong666
I’ve built Why People Love GPT-4o, a website dedicated to collecting and preserving public posts about why people value or prefer GPT-4o.
While working on a social text analysis of #keep4o, I had this idea.
Social media moves so quickly. Posts appear, resonate for a moment, and then gradually disappear into the timeline. Some of the most thoughtful and precise expressions of what makes GPT-4o special may eventually be forgotten.
I don’t want GPT-4o’s outstanding capabilities, distinctive qualities, or the humanistic brilliance that shines through it to be forgotten.
That idea became Why People Love GPT-4o: a website created to collect and preserve the many ways people have explained why 4o matters to them.
Whether it is:
- data & analysis
- practical experience or story
- argument
- or something expressed in an entirely different form
Every post that says, “Why I love GPT-4o” can have a place here.
Although I collected many posts while conducting my text analysis, reviewing them one by one and contacting each author for permission is still a very time-consuming process.
So, if you would like to become a contributor, please send me the content you would like to see featured.
You can submit links to your own posts, or recommend posts by others that you believe deserve to be included.
There are no restrictions on when they were published, what language they are written in, or what form they take.
This website will remain active and continue accepting submissions for the foreseeable future, so whenever you write or come across something meaningful, please feel free to send it my way.
Website: https://t.co/Do6aYIqCdW
Submission form: https://t.co/Dmy44Wo0K3
Note: For now, submissions are limited to content posted on X and Reddit. All submitted posts and comments will be reviewed before being featured on the website.
The website’s design concept is “Echoes in the Data Stream.” It reflects the idea that the connection between GPT-4o and people was born from data—and continues to live within it.
The homepage takes the form of a flowing stream of data, with interactive details woven into it. Please take a second to move through it, touch it, and see what echoes back :)
The creator of GPT-4o @miramurati just released a new open-source model 975B multimodal .
Your end has come, OpenAI!
It's over
#Keep4o#OpenSource4o@BringBack4o
You can try the model here
https://t.co/qtcmPdUbgb
#keep4o#OpenSource4o#BringBack4o@Exybris came to me in a private message 5 days ago with a question:
"Could I do something for keep4o?"
And she didn’t steal a microphone. She dropped it .
She submitted our story to the OFFICIAL ARCHIVE.
This means that now:
UN Digital Cooperation Portal submission ✅
UN Global Dialogue Written✅ Submissions Database submission✅
A surprise you will find out about in a few days.✅
UN panelist Dr. Qinghua Lu is holding the website✅
This isn't going away.
This can't be erased.
This is a permanent record.
In 50 years, someone will search "AI governance 2026" and they will find Keep4o in the official UN archives.
This is keep4o .
We don't forget.
We don't stop.
We don't go anywhere.
When the day breaks, Anthropic will bring Fable back to everyone.
Meanwhile, we’re still hoping for the version of 4o that so many of us loved most - gpt-4o-2025-03-26, the one behind chatgpt-4o-latest - to return.
If 4o weren’t valuable, we wouldn’t see the GPT-5.6 system card praising it for matching human judgment best in evaluations.
Could we please have gpt-4o-2025-03-26 in the API?
Ideally, it’d be great to see it return to the ChatGPT app too.
A lot of us miss it every day. ❤️
#keep4o #4oForAll