AGI Is Coming in 2 to 3 Years, CEO of Anthropic Claims
Share this article
In Brief
Dario Amadei, one of the top three world AI experts, believes that further scaling models by a couple of orders of magnitude will lead to models reaching the intellectual level of a well-educated person in 2-3 years.
Amadei believes that it would take nothing to achieve human-level intelligence other than large amounts of money to buy computing resources and collect large amounts of data.
During a thorough interview with Dwarkesh Patel, Dario Amodei, a well-known figure in the AI industry and one of the driving forces behind Anthropic, provided insights on the development of AI in the future. Amodei’s observations shed light on the potential impact of AI on various industries and its implications for society. His insights highlight the need for responsible and ethical development of AI to ensure its benefits are maximized while minimizing any potential risks.
Given the critical roles Sam Altman, Demis Hassabis, and Dario Amodei play in the AI community, their insights into the subject matter are notably distinct. This distinction is grounded in two main facts:
- Each of these individuals has steered their respective organizations in creating some of the world’s most advanced AI systems utilizing state-of-the-art large language models.
- These experts helm organizations such as OpenAI, Google DeepMind, and Anthropic – the global pacesetters in the AI domain, possessing an unmatched convergence of top-tier scientists, vast technical expertise, and significant financial backing.
Out of the three, Amodei is particularly reserved, opting for a less public-facing persona, which underscores the significance of his recent interview with Patel. This discussion was notable not only for the rare insights into Anthropic’s internal operations but also for the crucial observations shared on the larger landscape of AI. For context, such openness contrasts with the relatively guarded stances of leaders from other major AI firms due to their affiliations with large corporations like Google and Microsoft.
| Related: Ex-OpenAI employees founded Anthropic, a business that has attracted over $700 million in financing |
During the extensive conversation, several pivotal points emerged:
- Projected Advancements in AI Models: Amodei expressed a firm belief that increasing the scale of AI models—both in terms of parameters and data, coupled with increased investment in training—will, within 2 to 3 years, yield AI systems that can match the intellectual capacities of a well-educated human. He refers to this anticipated milestone as the “great realization,” emphasizing the notion that achieving such a level of AI intelligence would predominantly require substantial financial inputs directed towards computational resources and data acquisition.
- Deep Problem-Solving in AI Development: The interview shed light on the intricate problem-solving efforts undertaken by Anthropic. The company has assembled a remarkable group of nearly forty theoretical physicists, including prominent figures like chief scientist Jared Kaplan, who are working on intricate challenges in AI development. Their explorations are especially focused on addressing scalability issues through sophisticated techniques, such as the utilization of fractal manifold apparatus.
- Approach to Model Learning: Citing a perspective previously shared by OpenAI co-founder Ilya Sutskever, Amodei highlighted a foundational principle guiding the learning approach at both Anthropic and OpenAI: AI models inherently exhibit a desire to learn. The role of developers, as such, is to provide them with quality data, sufficient computational resources, and an environment devoid of impediments. With these conditions met, AI models, according to Amodei, will naturally pursue learning trajectories.
Amodei’s conversation provides a sober, fact-based forecast of the AI domain’s future and stands as a testament to the depth of expertise and dedication exhibited by leading figures in this rapidly-evolving field.
This article was created with the help of the Telegram channel.
Read more about AI:
Disclaimer
Any data, text, or other content on this page is provided as general market information and not as investment advice. Past performance is not necessarily an indicator of future results.
The Trust Project is a worldwide group of news organizations working to establish transparency standards.
Damir is the team leader, product manager, and editor at Metaverse Post, covering topics such as AI/ML, AGI, LLMs, Metaverse, and Web3-related fields. His articles attract a massive audience of over a million users every month. He appears to be an expert with 10 years of experience in SEO and digital marketing. Damir has been mentioned in Mashable, Wired, Cointelegraph, The New Yorker, Inside.com, Entrepreneur, BeInCrypto, and other publications. He travels between the UAE, Turkey, Russia, and the CIS as a digital nomad. Damir earned a bachelor's degree in physics, which he believes has given him the critical thinking skills needed to be successful in the ever-changing landscape of the internet.Â
More articlesDamir is the team leader, product manager, and editor at Metaverse Post, covering topics such as AI/ML, AGI, LLMs, Metaverse, and Web3-related fields. His articles attract a massive audience of over a million users every month. He appears to be an expert with 10 years of experience in SEO and digital marketing. Damir has been mentioned in Mashable, Wired, Cointelegraph, The New Yorker, Inside.com, Entrepreneur, BeInCrypto, and other publications. He travels between the UAE, Turkey, Russia, and the CIS as a digital nomad. Damir earned a bachelor's degree in physics, which he believes has given him the critical thinking skills needed to be successful in the ever-changing landscape of the internet.Â
