Musk After Calling for Halt of ‘Giant AI Experiments’ Is Reportedly Exploring His Own Through Twitter
Following the hiring of several former DeepMind researchers and the purchase of 10,000 GPUs, the social app is thought to be exploring a generative AI that will be trained on its massive in house data sets.
Just two weeks after Elon Musk and a thousand others called for the “pause of giant artificial intelligence experiments,” Twitter is said to be exploring its own, according to reports from Business Insider.
On March 29, the open letter published by the nonprofit Future of Life Institute expressed that AI labs are currently in an “out-of-control race” to develop and deploy machine learning systems “that no one — not even their creators — can understand, predict, or reliably control.” It goes on to call on all AI labs to “immediately pause for at least six months the training of AI systems more powerful than GPT-4,” expressing that “this pause should be public and verifiable, and include all key actors. If such a pause cannot be enacted quickly, governments should step in and institute a moratorium.”
Preferable to a moratorium however, the letter called on AI developers to work with policymakers to accelerate the development of robust AI governance systems, including regulatory authorities dedicated to AI, oversight, and tracking of highly capable AI systems and large pools of computational capability, auditing and certification ecosystem, liability for AI-caused harm, and institutions for coping with the economic and political disruptions AI may cause.
While many of the concerns are valid and regulation will certainly come into place, a large portion of the public responded that a pause would only stifle progress in the United States while other countries continue to advance research in the sector — which in his signature style, Musk posted a meme about.
Summary of argument against AI safety pic.twitter.com/Vmg4yJm22Y
— Elon Musk (@elonmusk) April 4, 2023
What of his giant experiment though? While details are scarce at the time, anything Twitter could be exploring can be considered to fit the classification of the very type of experiment the open letter called for a pause of. This is due to the massive amounts of data that the generative AI model will reportedly be trained on and the purchase of over 10,000 GPUs.
Musk is known for publicly saying one thing and doing another, be it for the sake of trolling or to cloak business dealings — a pattern that became evident during the Twitter acquisition saga. In this case, Musk’s support of the call to pause experiments could just be to give him time to shore up a stronger position in the sector.
The Chief Twit is also known for being competitive and could potentially be looking to launch a product rivaling OpenAI’s ChatGPT. Last month, Musk publicly shared criticisms for the AI research company he co-founded in 2015, tweeting “I’m still confused as to how a non-profit to which I donated ~$100M somehow became a $30B market cap for-profit. If this is legal, why doesn’t everyone do it?”
Hints that these criticisms may come from a place of a personal feud have also been made apparent by coverage from Seamfor, which highlighted the fact that in 2018, Musk was concerned about OpeanAI’s performance compared to Google and suggested he run the research company in place of co-founder Sam Altman — which both Altman and the other founders of OpenAI rejected.
Musk left not long after this rejection, and along with him any additional funding to the firm and its research. This departure left the company with a massive need to replace cash, pushing it to create a for-profit subsidiary in 2019, which then lead to a $1 billion USD investment from Microsoft.
Insider also reported that Musk was supposedly “furious” at the public launch of ChatGPT in November and that he restricted access to Twitter’s data, tweeting “I just learned that OpenAI had access to Twitter database for training. I put that on pause for now,” he added that “OpenAI was started as open-source & non-profit. Neither are still true.”
At the time of writing, no public statements have been made by Twitter or Musk regarding the potential AI project. That being said, it likely won’t be long before official word breaks, based on the CEO’s competitive nature, desire to control the narrative, and habit of tweeting whatever comes to mind.
Related, Alibaba Cloud unveils AI model Tongyi Qianwen, its answer to ChatGPT.