Google DeepMind Cofounder Envisions AI as Entrepreneur and Inventor in Next Five Years

Mustafa Suleyman, the cofounder of DeepMind, Google's AI division, predicts that AI will gain the ability to establish and operate its own business within the next five years.

During a Thursday panel on AI at the 2024 World Economic Forum, the current CEO of Inflection AI was questioned about the timeline for AI to pass a modern version of the Turing test. This test signifies that the technology has attained advanced, human-like capabilities referred to as AGI (artificial general intelligence).

In response, Suleyman suggested that the contemporary Turing test would assess whether AI can function as an entrepreneur, mini-project manager, and inventor, proficient in marketing, manufacturing, and selling a product for profit.

He anticipates that AI will demonstrate these business-oriented capabilities before 2030, and he envisions it happening inexpensively.

"I'm pretty sure that within the next five years, certainly before the end of the decade, we are going to have not just those capabilities, but those capabilities widely available for very cheap, potentially even in open source," stated Suleyman in Davos, Switzerland. "I think that completely changes the economy."

Suleyman's remarks form part of his broader predictions about the societal impact of AI, coinciding with the rise of tools like OpenAI's ChatGPT.

Earlier in the week, Suleyman asserted to CNBC in Davos that AI is a "fundamentally labor-replacing" tool in the long term.

In a previous interview with CNBC in September, he foresaw a future where everyone has AI assistants enhancing productivity and possessing an "intimate knowledge of personal information."

These AI assistants would be adept at reasoning over one's day, aiding in prioritizing time, fostering creativity, and more, according to Suleyman.

However, during the 2024 Davos panel, he expressed skepticism about using the term "intelligence" concerning AI, deeming it a "pretty unclear, hazy concept." He considers the term a "distraction."

Suleyman advocates for researchers to concentrate on AI's practical capabilities, focusing on whether an AI agent can effectively communicate with humans and undertake tasks such as planning, scheduling, and organizing.

He suggests stepping back from the "engineering research-led exciting definition" and, instead, directing attention to what AI can genuinely achieve.

As of now, Suleyman has not responded to further inquiries from BI via Inflection AI.

How was the article?

• Change Image

  Good
• Change Image

  Bad