paint-brush
OpenAI Sued by Elon Musk for Violating Non-Profit Charterby@legalpdf

OpenAI Sued by Elon Musk for Violating Non-Profit Charter

by Legal PDF: Tech Court CasesAugust 9th, 2024
Read on Terminal Reader
Read this story w/o Javascript
tldt arrow

Too Long; Didn't Read

In 2023, OpenAI, Inc. shifted from its initial open-access model to a more secretive approach. After promising open-source research, the organization began restricting access to its advanced models, including GPT-4, and exclusively licensed its technology to Microsoft. This move contradicts earlier commitments and has raised concerns about the privatization of technologies meant for public benefit.
featured image - OpenAI Sued by Elon Musk for Violating Non-Profit Charter
Legal PDF: Tech Court Cases HackerNoon profile picture

Elon Musk v OpenAI, Court Filing, retrieved on April 30, 2024, is part of HackerNoon’s Legal PDF Series. You can jump to any part of this filing here. This part is 10 of 29.

G. Defendants Renege in 2023

112. In its early years, OpenAI, Inc.’s research and development were performed in the open, providing the public with free access to designs, models, and code.


113. For example, in June 2018 when OpenAI, Inc. researchers discovered that an algorithm called “Transformers” could perform natural language tasks without any explicit training, entire communities from opensource, grass-roots groups to commercial endeavors sprung up to enhance and extend OpenAI, Inc.’s models—the intended benefit of making the non-profit’s research open source.


114. In 2019, OpenAI, Inc. released the full, open version of a secondgeneration Generative Pre-Trained Transformer (“GPT”), GPT-2 with the stated hope that it would “be useful to developers of future powerful models.” It also released a detailed report describing the new model and acknowledged some of the many benefits of openly releasing such models to the public.


115. In 2020, OpenAI, Inc. announced a third version of its model, GPT3 and again, published a research paper detailing its complete implementation for others to build on.


116. OpenAI, Inc.’s initial findings, while technologically interesting, had little commercial value and were openly published by Altman. But having reached the threshold of AGI, which under the founding agreement Defendants were to develop for the benefit of humanity rather than profit, Altman about-faced and began locking down the technology for personal gain.


117. For example, on March 14, 2023, OpenAI, Inc. released its most advanced model to date, GPT-4, which many including Microsoft celebrated as “a form of general intelligence.” Microsoft’s scientists stated that, given GPT-4’s advanced capabilities, “we believe [it] could reasonably be viewed as an early (yet still incomplete) version of an artificial general intelligence (AGI) system.” Defendants, however, publicly released no report or code, preventing the public from building on the non-profit’s AI advancements as Musk had been promised.


118. Reuters has reported that OpenAI is also developing a secret algorithm called Q*, and that several OpenAI staff members wrote a letter warning about its potential power. It appears Q* may be an even clearer and more striking example of AGI developed by OpenAI.


119. On information and belief, Altman caused the non-profit to exclusively license its technology to Microsoft, the world’s largest for-profit corporation,[6] contrary to Altman and the non-profit’s black-letter commitments—e.g., OpenAI, Inc.’s Certificate of Incorporation: “no part of the net income or assets of this corporation shall ever inure to the benefit of . . . any private person”; and Charter: “We commit to . . . avoid enabling uses of AI or AGI that . . . unduly concentrate power.”


120. The Microsoft license includes all OpenAI, Inc.’s “pre-AGI”[7] technologies, and tasks the Board with determining when “AGI” has been attained. To date, the Board has made no such finding, thus giving Microsoft unfettered access to OpenAI’s suite of technology. With full access to the nonprofit’s research and employees, many of whom, on information and belief, now work for the for-profit OpenAI enterprise or Microsoft, it is a short walk to develop a “complete” AGI system based on the non-profit’s research and technology.


121. Defendants have kept GPT-4, and subsequent models including without limitation, GPT-4T and GPT-4o (released May 2024), entirely closed. On information and belief, the internal details of GPT-4, GPT-4T, and GPT-4o are known only to OpenAI and its partner Microsoft. The reason for the secrecy is obvious: Defendants and Microsoft stand to make a fortune selling this technology to the public, which would not be possible if the non-profit made its research and technology freely available, as Altman had repeatedly promised Musk.



Continue Reading Here.


About HackerNoon Legal PDF Series: We bring you the most important technical and insightful public domain court case filings.


This court case retrieved on August 05, 2024, deadline.com is part of the public domain. The court-created documents are works of the federal government, and under copyright law, are automatically placed in the public domain and may be shared without legal restriction.

[6] As of August 2024, Microsoft is worth a reported $3.11 trillion.


[7] OpenAI’s website defines AGI as “a highly autonomous system that outperforms humans at most economically valuable work” and states such technology “is excluded from IP licenses and other commercial terms with Microsoft, which only apply to pre-AGI technology.”