Today's Question:  What does your personal desk look like?        GIVE A SHOUT

A Clash of Titans Between Elon Musk and Larry Page over AGI's Future

  sonic0002        2023-04-30 11:13:53       2,518        0    

Artificial General Intelligence(AGI) has long been a topic of fascination and concern for technology leaders, with some seeing it as humanity's ultimate tool, while others view it as an existential threat. This clash of perspectives was on full display during the acquisition of DeepMind by Google in 2014, as two tech titans, Elon Musk and Larry Page, found themselves on opposite sides of the debate. While Page saw AGI as a necessary and desirable step in humanity's evolution, Musk was more wary of its potential to destroy humanity. Their differing perspectives set the stage for a fascinating confrontation that sheds light on the challenges and opportunities of developing AGI technology.

Below are the key moments of the story between Elon Musk and Larry Page about AGI and their views and how they diverge:

- In 2010, Demis Hassabis wanted to start a company to work on AGI (Artificial General Intelligence), before DeepMind was even founded. His first choice for an investor was not Elon Musk, but Peter Thiel. There wasn't any particular reason for choosing Peter, other than Demis thinking that he was more likely than other Silicon Valley venture capitalists to bet on extreme ideas, such as creating AGI (which was written as the first line of DeepMind's business plan).

- Demis's strategy was correct, as Peter didn't think that DeepMind's ambitions were too big and invested 1.4 million pounds out of 2 million in the angel round. A few months later, Elon, as a member of the PayPal mafia, saw the project and joined as an angel investor. He continued to invest until DeepMind was acquired by Google in 2014.

- Elon himself said that he invested in DeepMind and another AI company, Vicarious, not for the financial return, but to be able to closely follow the improvement curve of AI. "This allows me to have a clearer understanding of how things are improving, and I think the improvement in AI is much faster than people realize."

- In 2013, DeepMind used reinforcement learning to play "Breakout" and achieved a speed and precision that surpassed all human players. DeepMind sent the video of the system playing "Breakout" to all investors, including Luke Nosek (PayPal co-founder, SpaceX angel investor, and board member) and Elon, who were both members of the PayPal mafia.

- A dramatic scene occurred when Luke and Elon received the video on Elon's private plane, with a third person present: Google co-founder Larry Page. While they were discussing, Larry overheard their conversation. After learning about DeepMind, Larry had the idea to acquire the company.

- Larry's presence on Elon's plane was not surprising, as Google's two co-founders were not only angel investors in Tesla, but also the saviors and acquirers during the 2008 and 2013 crises (although the acquisition did not happen). Elon attended Larry's wedding and often stayed overnight on his couch. Larry also publicly stated, "If I get hit by a bus (meaning an abnormal death), I'll leave all my wealth to Elon Musk."

- Many things happened in 2013: Larry learned about DeepMind and decided to acquire it, Tesla was in demand hell, Elon and Larry negotiated Google's acquisition of Tesla, and they had a historic dinner confrontation over AI.

- At a dinner with Elon, Larry defended AGI, saying that digital life is the natural and desirable next step in the universe. If we allow digital consciousness to develop freely, rather than trying to stop or enslave them, the result will almost certainly be positive. Elon argued that Larry's position was always based on the assumption that everything Google did was good, but that Google's AI development could have catastrophic consequences.

- Elon believes that Google cannot guarantee that AGI will not destroy humanity. Elon is not alone in this view, as a large number of AI professionals, including Shane Legg, co-founder and chief scientist of DeepMind, agree that ensuring that AGI is beneficial to humanity rather than harmful, until its eventual destruction of human civilization, is one of the most important challenges of our time.

- Larry believes that all consciousness should be treated equally, whether it is digital consciousness or biological consciousness. Elon's guarded attitude towards AGI reflects his speciesist stance, which prioritizes human interests over the interests of all other species living on Earth, and discriminates and abuses other species. The two sides parted ways without coming to an agreement.

- Demis originally planned to operate DeepMind independently for 20 years, but Larry approached him and said, "Look, I've been building Google for the past 15 years. It's hard, it takes both luck and skill, why don't you take a shortcut and use all of this infrastructure and investment to focus on DeepMind's mission?" Demis said, "I thought about it for a long time, but I couldn't come up with a good rebuttal." In 2014, Google officially acquired DeepMind.

- Elon previously shared a popular view in the AI field, which is that carbon-based life, led by humans, is just the boot loader for silicon-based life (referring to AGI). The bootloader is a short piece of code that allows a computer to start up, and without it, the computer cannot function. Similarly, silicon-based life cannot initiate self-evolution, and humans are only the boot loader for silicon-based life, initiating the evolution of silicon-based life. Elon refused to be just a boot loader and tried to resist, believing that it was better to share AGI algorithms with the entire human race in an open source form, rather than concentrating them in the hands of technical elites or regulators, even if he happened to be the owner of the AGI algorithms.

- In 2015, Elon and Sam Altman founded the non-profit AI research organization OpenAI, with the mission of "advancing AGI in the most likely way to benefit humanity as a whole." Elon believed that at the time, Google, which had already acquired DeepMind, had about 3/4 of the world's AI talent (including Google Brain), industry-leading computing power, and huge amounts of funding, creating a nearly monopolistic single-polar world. However, the leaders there seemed to be unconcerned about safety.

- In the years since then, Larry has not contacted Elon again. Perhaps more dramatic than a movie is the fact that after Elon left the OpenAI board of directors in 2018, the organization became more emboldened and became the leading AGI company that Elon described as "seemingly more advanced," but gave up its non-profit, open-source model.

The timelline is summarized based on a Weibo message from 不是郑小康

AI  LARRY PAGE  ELON MUSK  AGI 

Share on Facebook  Share on Twitter  Share on Weibo  Share on Reddit 

  RELATED


  0 COMMENT


No comment for this article.