Activities
Divisions
Performances
Activities
Divisions
Performances
NVIDIA's Jensen Huang believes AI hallucinations can be fixed and anticipates artificial general intelligence within 5 years
Jensen Huang, the head of NVIDIA, predicts that AGI, or artificial general intelligence, will be a reality much sooner than 2030. He even suggests that we could witness the first AGIs in the coming 5 years.
AGI, also known as Artificial General Intelligence, is a hot topic in the AI industry and a significant achievement that nearly all AI professionals currently anticipate to be on the horizon. NVIDIA's CEO, Jensen Huang, suggests that AGI will be a reality in the near future.
AGI is projected to bring about a significant advancement in technological prowess. Also known as "strong AI" or "human-level AI", AGI implies the possibility of machines displaying cognitive skills similar to or even better than human beings. In contrast to standard or narrow AI, that focuses on particular tasks, AGI is expected to be proficient in a broad range of cognitive areas.
During Nvidia's yearly GTC developer conference, CEO Jensen Huang spoke to the media, providing his perspective on the direction of AGI and dealing with the profound questions it brings up. Although Huang recognized the importance of AGI, he also showed a hint of tiredness from the constant questioning on the subject, blaming this exhaustion on the media's regular misunderstanding of his comments.
The rise of Artificial General Intelligence (AGI) raises deep existential questions, challenging human dominance and purpose in a future where machines could exceed human abilities. At the heart of these worries is the uncertainty of AGI's decision-making methods and goals, which might deviate from human ethics and priorities – a concept that has been a subject of science fiction for many years.
While some media sources are keen on obtaining a schedule for AGI's progression, Huang stressed the difficulty in pinning down what AGI exactly is and warned against over-the-top conjecture. By comparing it to concrete achievements such as New Year's Day or arriving at a destination, Huang highlighted the need for agreement on the standards used to determine AGI achievement.
Huang presented a detailed view, proposing attainable goals for AGI, and indicating a five-year timeline for certain performance standards. Nevertheless, he stressed the importance of precisely outlining AGI's boundaries to make precise forecasts.
Speaking on the issue of AI hallucinations – scenarios where AI produces believable but incorrect replies—Huang suggested a solution based on comprehensive investigation. He introduced a "retrieval-augmented generation" method, similar to fundamental media literacy, where AI cross-checks responses with trustworthy sources before giving an answer. Huang emphasized the importance of corroborating multiple sources, especially in crucial areas like health advice, to maintain accuracy.
Fundamentally, Huang's observations illuminate the intricate nature of AGI creation and the crucial need for accountable AI management to lessen possible dangers. With the ongoing progression of AI, all parties involved must tackle moral issues and implement plans to guarantee that AI systems correspond with human principles and prioritize the welfare of society.
Search for us on YouTube
Highlighted Programs
Associated Reports
John Mearsheimer advises Palki Sharma that the US ought to cease aid to Ukraine and assures that China won't invade Taiwan
OpenAI's Sora has the ability to create life-like nude videos, prompting developers to hasten their efforts to rectify this
At last, Apple unveils its MM1, a multimodal AI model that generates text and images
China advises the US to refrain from getting involved in the South China Sea dispute
John Mearsheimer informs Palki Sharma that the US must discontinue its support to Ukraine and reassures that Taiwan won't be attacked by China
OpenAI's Sora possesses the capability to produce nude videos that appear realistic, causing developers to rush to find a solution
Finally, Apple introduces its MM1, an artificial intelligence model that's multimodal and generates both text and images
China warns the US to avoid meddling in the South China Sea controversy
is available on YouTube
Firstpost owns all rights and permissions, protected by copyright, as of
+ There are no comments
Add yours