[mindcrime]

There are a handful of "open source LLM" initiatives out there, although I don't think any of them are quite up to the level of ChatGPT. Possibly one of the more interesting ones is GLM-130B.

https://github.com/THUDM/GLM-130B

Released by some folks at Tsinghua University in China, back in August. The model itself is licensed under some janky "free to use, but not open source" license, but it looks like most of the code for training, evaluation, etc. is available and licensed under either the Apache License or a BSD-like license.

You might also find this of interest:

https://arxiv.org/pdf/2103.08894 - "Distributed Deep Learning Using Volunteer Computing-Like Paradigm"

FWIW, I tend to agree with your overall sentiment. As AI becomes progressively more capable, it represents an ever increasing possibility of consolidating more and more power into the hands of fewer and fewer entities. I believe that one way to counter that (albeit not one without its own risks) is to democratize access to AI as much as possible.

Actually, now that I think about it, wasn't something along those lines purportedly the original idea behind OpenAI in the first place? Or am I having a "Mandela Moment" and mis-remembering?

[yummytummy]

Thank you, I will take a look at the links! Licensing might be an issue, but as long as the models and checkpoints are available there's at least competition.

I also seem to remember that this was the original intent of OpenAI, but me confirming that without a source only adds one more data point to the Mandela Effect ;).

ChatGPT says "The original mission of OpenAI was to advance artificial intelligence in a way that is safe and beneficial for humanity". This seems like a demonstration of the point I'm trying to make.

[RGamma]

Concerning your last paragraph see https://openai.com/charter/

[yummytummy]

Thank you! There's a lot of talk about "safety" here that I think back in 2018 I interpreted as guard rails towards some sort of rogue AGI.

"and to avoid enabling uses of AI or AGI that harm humanity or unduly concentrate power."

The current restrictions seem like the opposite of this goal.

[mindcrime]

Just for kicks and giggles, here's what ChatGPT has to say:

Q:

Why should advanced Artificial Intelligence be made open source to ensure democratization of access and avoid undue concentration of power?

A:

There are a few potential reasons why some people might argue that advanced artificial intelligence (AI) should be made open source in order to promote democratization of access and avoid concentration of power:

    Access to advanced AI technologies can be a key determinant of a company's or country's competitiveness and prosperity. If only a few organizations have access to these technologies, they may have a significant advantage over others. Making advanced AI open source could help to level the playing field and ensure that more organizations have the opportunity to develop and use these technologies.

    Making advanced AI open source could help to promote the development of a more diverse and vibrant ecosystem of AI innovation. By enabling more organizations and individuals to access and build upon advanced AI technologies, it could encourage a greater diversity of approaches and ideas, potentially leading to more rapid progress and innovation in the field.

    Open sourcing advanced AI could help to ensure that these technologies are developed and used in a way that is transparent, accountable, and aligned with the values and goals of society. By making the underlying algorithms and data available for scrutiny, it could be easier to ensure that AI systems are fair, unbiased, and do not have unintended consequences.

However, it's also important to note that there are potential downsides to open sourcing advanced AI, such as the risk of intellectual property theft or the possibility that the technology could be used for malicious purposes. As with any complex issue, there are likely to be trade-offs involved in the decision of whether or not to open source advanced AI.