The Association for Human Mathematics (AHM) on Thursday, alleged that OpenAI was guilty of plagiarism, copyright infringement and and trademark violation and urged other mathematicians not to work with the company after it released 722 manuscripts spanning 372 groups of mathematical results.
The firm said that the documents included solutions to hundreds of open problems in the field of mathematics and theoretical computer science. OpenAI posted these results within a Github repository along with protocols for citiations and review of their papers.
ALSO READ: OpenAI's GPT-6 Brings Intelligent UI To ChatGPT; Free Users Get Access From October 8
OpenAI stated that it had consulted with the Advisory Group on Mathematics and Artificial Intelligence (AGMAI) at the Institute for Advanced Study in order to refine the manner in which it shares its mathematical results with the wider public.
The AHM criticised OpenAI, alleging that it ignored AGMAI's own initial advisory statement which said that leading AI companies should not use their internal AI models to evaluate advanced mathematical problems. It further argued that as per the Advisory Group's recommendations using a private, proprietary model to solve complicated mathematical problems creates a two-tiered system where a small number of labs make breakthroughs which alienates the wider mathematical community that cannot keep pace.
There is also the question of transparency as AI models cannot show how their work in order to demonstrate how they got these results or independently recreate their work. This process also does not include open publication, peer review and independent verification. The company received these criticisms regarding transparency when it claimed to have solved the Naiver-Stokes mathematics problem.
“We reject OpenAI's assertion that this release advances our subject, and we urge mathematicians and the public to view the value of this publication model with due skepticism," the association said in its public statement.
It accused the company of "total disregard for the norms of scientific research", Norms which it described as "trustworthy, ethically researched, and in the public interest."
ALSO READ: Google Launches Workplace AI Agent That Acts Like Colleague
It further stated that the ChatGPT parent company releasing these many manuscripts is not a demonstration of scholarship, but of power.
We urge mathematicians to discontinue their work with OpenAI and to return to a vision of science that centres human understanding," the organisation said.
Essential Business Intelligence, Sharp Market Insights, Practical Personal Finance Advice, Daily Fuel, Gold and Silver Prices and Latest Stories — On NDTV Profit.