>It’s not that China’s AI is “surpassing the US,” but rather that “open source models are surpassing proprietary ones,” LeCun said in a post on Threads.
>DeepSeek’s R1 is itself open source, as is Meta’s Llama. OpenAI, which was originally founded as an open-source AI company with a mission to create technology that benefits all of humanity, has on the other hand more recently shifted to closed-source.
>LeCun said DeepSeek has “profited from open research and open source.”
>”They came up with new ideas and built them on top of other people’s work. Because their work is published and open source, everyone can profit from it,” LeCun said. “That is the power of open research and open source.”
shadowrun456 on
Open source is better than closed source.
In other news, water is still wet. More at 11.
Crivos on
What? Humans looking out for each other yields positive results? Shocking.
Alternative-End-8888 on
Hard to believe Chinese I.P. as breakthrough is Open Source, definitely a Sputnik Moment on USA..
Moreso than these hypersonic or sixth gen Hollywood productions of China propaganda.
the_meat_fest on
None of them have a moat… Especially not against open source.
zethuz on
I hope there are lessons in humility from this achievement, especially for Sam Altman
2roK on
Excuse me? I thought we agreed that the AI world dominance should go to Muskrat, Bezos and Robotberg?
dustofdeath on
Because open models do not have to follow some sanitized safety rules.
So they are more open if anyone can train them on any content (which can also be bad).
A model that is supposed to generate accurate humans. Can’t be accurate if it is censored because of nudity – it lacks accurate anatomical data.
8 Comments
From the article
>For Meta’s chief AI scientist, Yann LeCun, the biggest takeaway from DeepSeek’s success was not the heightened threat posed by Chinese competition but the value of keeping [AI models open source](https://www.businessinsider.com/china-startup-deepseek-openai-america-ai-2025-1) so that anyone can benefit.
>It’s not that China’s AI is “surpassing the US,” but rather that “open source models are surpassing proprietary ones,” LeCun said in a post on Threads.
>DeepSeek’s R1 is itself open source, as is Meta’s Llama. OpenAI, which was originally founded as an open-source AI company with a mission to create technology that benefits all of humanity, has on the other hand more recently shifted to closed-source.
>LeCun said DeepSeek has “profited from open research and open source.”
>”They came up with new ideas and built them on top of other people’s work. Because their work is published and open source, everyone can profit from it,” LeCun said. “That is the power of open research and open source.”
Open source is better than closed source.
In other news, water is still wet. More at 11.
What? Humans looking out for each other yields positive results? Shocking.
Hard to believe Chinese I.P. as breakthrough is Open Source, definitely a Sputnik Moment on USA..
Moreso than these hypersonic or sixth gen Hollywood productions of China propaganda.
None of them have a moat… Especially not against open source.
I hope there are lessons in humility from this achievement, especially for Sam Altman
Excuse me? I thought we agreed that the AI world dominance should go to Muskrat, Bezos and Robotberg?
Because open models do not have to follow some sanitized safety rules.
So they are more open if anyone can train them on any content (which can also be bad).
A model that is supposed to generate accurate humans. Can’t be accurate if it is censored because of nudity – it lacks accurate anatomical data.