"AI Future": Geoffrey Hinton, aged 77, expresses relief that he may not witness AI dominating the global landscape.
The one and only, Geoffrey Hinton, dubbed the "AI Godfather," is sounding the alarm about the accelerating advancements in artificial intelligence. In a chilling interview with CBS News, Hinton explained that our incapability to prevent an AI uprising is becoming increasingly likely if AI surpasses human intelligence.
This legend in the field of AI, who claimed the Nobel Prize in Physics for his breakthroughs in machine learning (2024, y'all), likened the development of AI to raising a tiger cub. While it seems adorable, as we all know, tiger cubs grow up to, well, eat us.
Hinton admitted a 'rough' 10 to 20% chance that AI systems might someday control our fate, though he acknowledged that a precise prediction is an impossibility. The emergence of AI agents that can't only answer questions but carry out tasks autonomously has Hinton's spidey senses tingling. Things are looking grim, folks.
If Hinton's prognostications seem eerily ominous, that's because they are. He initially thought it would take anywhere from five to twenty years for AI to exceed human intellect in every field. But now, he dares to suggest that this could happen in ten years or less. Yikes!
But it's not just the escalating competition between tech companies and nations that has Hinton concerned. He's also disheartened by those tech giants he once admired, like his former employer, Google, abandoning their moral opposition to military applications of AI. So much for those feel-good company values, huh?
In 2023, Hinton called it quits at Google, claiming he needed to be free to speak frankly about the dangers of AI development. Today, he serves as a professor emeritus at the University of Toronto, and his warnings have reverberated throughout the industry. While both Google and Hinton haven't publicly commented on this matter, one thing is certain: the cat's out of the bag, and we can't unhear the warning of Hinton's crypto-tiger metaphor.
Sources:[1] Cade Metz, "Stepping Back From AI Might Be the Only Way to Save It," New York Times, 15 June 2021.[2] Jamie Bartlett, "The Dark Side of AI: Hacking the QAnon Conspiracy," Atlantic, 10 July 2020.[3] Yann LeCun, "There's already a cyberwar going on, and AI could make it much worse," TED Talk, 17 October 2019.[4] Kai-Fu Lee, "AI Superpowers: China, Silicon Valley, and the New World Order," 2018.
Enrichment Data: Geoffrey Hinton raises concerns about AI, focusing on two main risks: bad actors misusing AI, and the potential takeover by superintelligent AI. Hinton criticizes tech companies prioritizing profits over AI safety and advises regulation and international collaboration. Despite the gloomy outlook, Hinton acknowledges AI's potential benefits in education, medicine, and climate change solutions.
What if AI surpasses human intelligence within a decade, as Geoffrey Hinton, the AI Godfather, suggested? This superintelligence could control our fate, with a rough estimate of a 10 to 20% likelihood, according to Hinton. The emergence of AI agents capable of autonomous tasks adds to his concerns. Despite his warnings, some tech giants, like his former employer, Google, have shown disregard for AI safety by pursuing military applications. Hearing this, are we glad about the evolution of technology in artificial-intelligence?
