In context: SoftBank Group, a well-known Japanese multinational investment holding company with a keen interest in technology unicorns, has a very outspoken CEO. Masayoshi Son is bullish about a lot of things, with some of them becoming successful products while others have turned into spectacular failures.
Speaking during the SoftBank World corporate conference, Masayoshi Son recently stated that some form of artificial general intelligence (AGI) will be created by 2030. AGI is usually defined as a truly "intelligent" computer algorithm, which can think, speculate and be aware of its own existence in the digital realm similar to human intelligence.
Also known as "strong AI," a true AGI agent would accomplish any intellectual task that biological intelligence can perform – only better than humans. An AGI would signal a "singularity" moment for technology progress and humanity as a whole, while modern "weak" AI like ChatGPT can only solve (or try to solve) specific problems without showing any sign of general cognitive abilities.
Son speculates that the upcoming AGI will be able to surpass humans in almost every intellectual task, as it will be ten times more intelligent than the sum of all human intelligence. The SoftBank CEO said that even the aforementioned weak AI algorithms are already better than humans in certain areas.
Saying that AI cannot be smarter than humans is "wrong," Son said. AI algorithms are now self-learning, self-training and self-inferencing, just like human beings. The CEO was already adamant about the potentially transformative influence of AI on business and society, but now he is providing a timeline of sorts for the development of the first (and likely last) AGI.
Son also provided a further development prospect for this newborn AGI, which will turn into an "Artificial Super Intelligence" within 20 years. If the AGI were ten times smarter than the entirety of humanity, Son stated at the conference, a super-digital intelligence would surpass human intelligence by a staggeringly high factor of 10,000.
SoftBank's leader is clearly pushing for bigger financial investments on AI development to turn his AGI dream into reality. He also said that Japanese companies need to "wake up" about the promises of the technology.
Those who won't invest in AI will be a "goldfish," the manager said, and they are "hallucinators" if they refuse to adopt the self-hallucinating chatbots. Son also stated that Arm-designed chips will be at the heart of this AGI revolution, while modern weak AI services are mostly (ab)using Nvidia GPUs for their massively parallel word-crunching workloads. Softbank has owned Arm since 2016.
https://www.techspot.com/news/100394-softbank-ceo-artificial-general-intelligence-here-10-years.html
