403
Sorry!!
Error! We're sorry, but the page you were looking for doesn't exist.
Experts Warn of “Catastrophic Outcomes” as AGI Approaches
(MENAFN) The journey toward developing artificial general intelligence (AGI) may carry risks of “catastrophic outcomes,” including cyberattacks on critical energy or water systems, according to Google DeepMind CEO Demis Hassabis.
He suggested that AGI could emerge within the next ten years.
Speaking at the Axios AI+ Summit in San Francisco last week, Hassabis defined AGI as a model that demonstrates “all the cognitive capabilities” of humans, encompassing inventive and creative skills.
He noted that current large language models are still “jagged intelligences,” with noticeable gaps in reasoning, long-term planning, and continuous learning. Nevertheless, Hassabis indicated that AGI might soon become achievable with continued scaling and “one or two more big breakthroughs.”
At the same time, he recognized that the lead-up to AGI is likely to involve tangible hazards and “catastrophic outcomes,” including cyberattacks targeting energy or water infrastructure.
“That’s probably almost already happening now… maybe not with very sophisticated AI yet,” he remarked, identifying this as the “most obvious vulnerable vector.”
Hassabis emphasized that the threat posed by malicious actors, autonomous agents, and systems that “deviate” from their intended objectives demands serious attention. “It’s non-zero,” he warned, referring to the possibility that advanced systems could “jump the guardrail.”
Hassabis’ warnings align with broader concerns within the tech sector. An open letter released in October and endorsed by leading technologists and public figures cautioned that “superintelligent” systems might endanger human freedom or even survival.
The letter calls for a worldwide halt on AI development until safety measures are verified. Among the signatories are Apple co-founder Steve Wozniak, AI pioneers Geoffrey Hinton and Yoshua Bengio, Virgin Group founder Richard Branson, along with influential political and cultural figures.
He suggested that AGI could emerge within the next ten years.
Speaking at the Axios AI+ Summit in San Francisco last week, Hassabis defined AGI as a model that demonstrates “all the cognitive capabilities” of humans, encompassing inventive and creative skills.
He noted that current large language models are still “jagged intelligences,” with noticeable gaps in reasoning, long-term planning, and continuous learning. Nevertheless, Hassabis indicated that AGI might soon become achievable with continued scaling and “one or two more big breakthroughs.”
At the same time, he recognized that the lead-up to AGI is likely to involve tangible hazards and “catastrophic outcomes,” including cyberattacks targeting energy or water infrastructure.
“That’s probably almost already happening now… maybe not with very sophisticated AI yet,” he remarked, identifying this as the “most obvious vulnerable vector.”
Hassabis emphasized that the threat posed by malicious actors, autonomous agents, and systems that “deviate” from their intended objectives demands serious attention. “It’s non-zero,” he warned, referring to the possibility that advanced systems could “jump the guardrail.”
Hassabis’ warnings align with broader concerns within the tech sector. An open letter released in October and endorsed by leading technologists and public figures cautioned that “superintelligent” systems might endanger human freedom or even survival.
The letter calls for a worldwide halt on AI development until safety measures are verified. Among the signatories are Apple co-founder Steve Wozniak, AI pioneers Geoffrey Hinton and Yoshua Bengio, Virgin Group founder Richard Branson, along with influential political and cultural figures.
Legal Disclaimer:
MENAFN provides the
information “as is” without warranty of any kind. We do not accept
any responsibility or liability for the accuracy, content, images,
videos, licenses, completeness, legality, or reliability of the information
contained in this article. If you have any complaints or copyright
issues related to this article, kindly contact the provider above.
Most popular stories
Market Research
More Story
Comments
No comment