Was OpenAI Brouhaha Aimed To Derail AI Capability To Complete Grade School Math?
It is being reported that a group of staff researchers at OpenAI conveyed their concerns to the board of directors regarding a significant artificial intelligence (AI) breakthrough, as reported by two individuals familiar with the matter to Reuters, which may be partly related to CEO Sam Altman’s removal from the executive role last week.
The undisclosed letter and the unveiling of an AI algorithm is said to have played a pivotal role leading up to Altman’s removal from the board. Altman, recognized as a prominent figure in generative AI, faced various issues raised by the board, including worries about advancing technology without a thorough understanding of its potential consequences. Over 700 employees had expressed their intent to resign in solidarity with Altman before his eventual reinstatement on Tuesday.
READ: Much Ado About Nothing: Sam Altman Back As OpenAI CEO, But At What Cost?
The sources indicated that the letter was one of several factors contributing to Altman’s dismissal. The researchers who authored the letter did not provide comments in response to requests.
Following contact from Reuters, OpenAI, while refraining from commenting, internally acknowledged the existence of a project referred to as Q* (pronounced Q-Star) and a letter to the board preceding the recent events. An OpenAI spokesperson mentioned that an internal message from Mira Murati, a longstanding executive, informed staff about certain media reports without verifying their accuracy.
Some members within OpenAI view Q* as a potential breakthrough in the pursuit of artificial general intelligence (AGI), which surpasses human capabilities in most economically valuable tasks. The sources suggest that, utilizing extensive computing resources, Q* demonstrated proficiency in solving specific mathematical problems. Although currently performing at a level comparable to grade-school students, the success in such tests has fostered optimism among researchers regarding Q*’s future potential.
Reuters could not independently verify the purported capabilities of Q* as claimed by the researchers.
Researchers consider mathematics a frontier in generative AI development, with current AI capabilities excelling in tasks such as writing and language translation. The ability to solve mathematical problems, where there is a singular correct answer, suggests greater reasoning capabilities akin to human intelligence. Such advancements could be applied to novel scientific research, according to AI researchers.
In the letter to the board, researchers highlighted the prowess and potential dangers of AI without specifying the exact safety concerns. The longstanding debate within the computer science community about the potential dangers posed by highly intelligent machines, including the possibility of autonomous decision-making detrimental to humanity, was underscored.
The existence of an “AI scientist” team was also flagged by researchers, with multiple sources confirming its formation by merging the “Code Gen” and “Math Gen” teams. This team focused on optimizing existing AI models to enhance reasoning and eventually undertake scientific work.
Altman, who played a key role in the rapid growth of ChatGPT, secured investments and computing resources from Microsoft to advance towards AGI. In a recent demonstration, Altman announced new tools and hinted at major AI advances during a summit in San Francisco. However, a day after his remarks, Altman was dismissed by the board.
Information for this briefing was found via Reuters and the sources mentioned. The author has no securities or affiliations related to this organization. Not a recommendation to buy or sell. Always do additional research and consult a professional before purchasing a security. The author holds no licenses.
