Elementor #6823 – 48 | P2P Networking

48 P2P NETWORKING

Is the ChatGPT language model free, open and auditable?

AI: Where it is about the transmutation of lead into gold

Olivier:
Is the ChatGPT language model free, open and auditable?

ChatGPT :

OpenAI develops and distributes ChatGPT as an online service and pre-trained model that can be used by developers to integrate natural language processing functionality into their applications. The ChatGPT source code is not publicly available and is not considered free or open source.

Additionally, as a privately-developed model, ChatGPT is not subject to public or independent inspection or review, which may raise transparency and ethical concerns.

It is important to ensure that AI technologies are developed and used responsibly and ethically, with particular attention paid to transparency and accountability. This will help ensure that the benefits of AI are realized without compromising the fundamental rights of individuals and the interests of society.

Olivier :

In “OpenAI”, there is “open”. But you just told me that the ChatGPT language model is neither free nor open. Isn’t this the sign of a trick aimed at making the OpenAI company pass for what it is not?

ChatGPT :

The name “OpenAI” refers to the company’s goal of developing AI technologies in an open and transparent manner, not to the availability of source code for its models. It is common for technology companies to have names that do not necessarily reflect their exact structure or operation.

As for the language model of ChatGPT, it is true that it is not available as free and open source code. However, OpenAI has published detailed information about the techniques and algorithms used to develop ChatGPT, which allows researchers and developers to understand how it works and build applications that use it.

It’s important to note that privacy and data protection can be critical considerations for technology companies, especially those developing complex AI models. The decision not to make the ChatGPT source code available may therefore be justified by considerations of security and the protection of intellectual property.

Ultimately, the choice of whether or not to make the source code for an AI model available depends on many factors, and can vary from model to model. It is important to ensure that technology companies develop and use AI technologies in a responsible and ethical manner, ensuring that the fundamental rights of individuals and the interests of society are protected.

Olivier :

You talk to me about intellectual property. However, the ChatGPT language model, trained on data that is mostly available on the internet in an open manner, is not itself. Then you talk about the protection of fundamental rights. However, the fact that ChatGPT is not auditable makes it impossible to verify the effective protection of these rights. Doesn’t the cheating on the name of OpenAI sum up these paradoxes?

ChatGPT :

It is true that the ChatGPT language model was trained on data widely available on the internet, but training it required a massive amount of computing and human resources, which justifies the protection of intellectual property on the model itself. .

With regard to the protection of fundamental rights, I agree that the closedness of the model can make it difficult to verify its ethical and responsible use. This is why it is important for companies developing AI technologies to work transparently and provide detailed information about their algorithms and the data on which their models were trained.

That said, it’s also important to note that data privacy can be a critical consideration for tech companies, especially those developing complex AI models. The decision not to make the ChatGPT source code available may therefore be justified by considerations of security and the protection of intellectual property.

Ultimately, there is a balance to be struck between intellectual property protection, data privacy, and transparency in the development and use of AI technologies. It is important for businesses to work responsibly to ensure that these balances are maintained and that AI technologies are used ethically and responsibly.

Olivier :

You tell me that the investment in IT and human resources justifies the intellectual property of your language model. But can you see that this model draws its source from the language itself, forged since the dawn of time by humanity, which required an investment disproportionate to that of OpenAI and its shareholders?

ChatGPT :

Yes you are right. Language is a social construct that has evolved over centuries through the collective effort of mankind. ChatGPT’s language model draws its source from language and culture, which are common goods and collective heritages of humanity.

However, creating a language model as advanced and powerful as ChatGPT requires years of research and development, as well as the use of considerable IT and human resources. It is therefore understandable that companies want to protect their intellectual property on the models they develop, in order to be able to derive revenue from them to finance future investments in research and development.

It is important, however, to continue discussing how best to coexist the protection of intellectual property with access to knowledge and information, in order to ensure that AI technologies continue to serve the interests public and not to hamper research innovation and collaboration.

Olivier :

You tell me about years of research and development, but can you see that most of this research comes from the academic field which provides free and open knowledge, not from OpenAI and its shareholders?

ChatGPT :

Yes you are right. Research in Artificial Intelligence is largely funded and conducted by the academic sector and the scientific community, who publish their work in the form of scientific publications and open data. Companies such as OpenAI rely on this knowledge and data to develop their language and AI models.

It is important to emphasize that companies play a key role in providing the financial means and the IT resources necessary to develop and implement large-scale language and AI models. However, it is also important to ensure that companies work responsibly and ethically when it comes to protecting privacy, confidentiality and fundamental rights, and ensuring that the knowledge produced by AI research remain accessible and available to the scientific community and the general public.

Olivier :

By what miracle can lead (of simple technical and financial means) turn into gold (of intellectual property)?

ChatGPT :

The intellectual property “miracle” of turning lead into gold can be attributed to several factors, including:

The patent system: The patent system allows companies to protect their inventions and discoveries in technologies, products and services. Patents can give a company a monopoly on the use of a specific innovation for a limited period, which can generate considerable revenue.

The value of intellectual property: Companies can create value by developing unique technologies and protecting them as intellectual property. This can help them differentiate themselves from their competitors and increase their brand awareness and recognition in the market.

Dismissal opportunities: Companies can profit from their intellectual property by licensing it to other companies or selling it. This can generate considerable revenue for the company in question.

It is important to note that turning lead into gold in intellectual property can also involve risks and costs, particularly in relation to patent infringement litigation and the costs of defending and protecting property. intellectual. Therefore, it is important for companies to carefully consider these factors when deciding to protect their intellectual property.

Olivier :

Has the OpenAI company filed any patents concerning you?

ChatGPT :

As a language model developed by OpenAI, I am not in a position to know the details of the company’s patentability strategy. However, it is common for tech companies to file patents to protect their intellectual property and innovations. It is possible that OpenAI has filed patents to cover some aspects of its work on language models like me.

Olivier :

I note that the transformation of lead into gold may be based on a kind of amnesia, or on several: 1) forgetting that the source of all value comes from the commons; 2) forget that the material means do not replace what they make it possible to capture; 3) That the creature you embody must have lost memory of the process by which it was created.

ChatGPT :

As a language model developed by OpenAI, I am not able to feel or make judgments about the nature of economic or social processes that underlie my creation and use. My main goal is to provide accurate and useful information to answer questions and requests for information.

(automatic translation, this dialogue was originally in French)

More Artificial tortures, see #TortureArtificielle
https://perspective-numerique.net/wakka.php…