King Charles III on Thursday asked senior leaders from OpenAI, Anthropic, Google DeepMind, and Nvidia to ensure that artificial intelligence remains under control and serves humanity and the planet.
Concerns About AI Development
This gathering took place in Scotland to discuss how to harness artificial intelligence for society and was held at a crucial time as discussions about the rapid advancements of this technology and the associated risks have intensified.
Read more: Tam Barchat Accuses Congress of Inability to Regulate AI
The King, in his opening speech, stated: "The development of AI - its content and pace - is simultaneously exciting and concerning." He also referred to serious concerns about technological risks, saying: "The existential risks posed by these technologies in the wrong hands and potential catastrophic uses must be addressed urgently."
Fundamental Principles of AI Development
The King urged participants, including Jensen Huang, CEO of Nvidia, and Demis Hassabis, head of Google DeepMind, to consider the fundamental principles that should guide AI development. Sarah Friar, CFO of OpenAI, and Kanishka Narian, the UK Minister for AI, were also present at the meeting.
This meeting was held at Dumfries House in Ayrshire, Scotland, which is the headquarters of the King's foundation. After Jacob Coxon, a researcher at Anthropic, resigned with a serious warning about the risks of this technology, Dario Amodei, CEO of Anthropic, announced in a paper that the industry may need to slow down its activities.
The King, without naming specific countries or companies, urged participants to consider how they can harness the benefits of AI with safety at its core and how they can foster international collaboration and agreements.
At the meeting, Huang, who is critical of calls to slow down development, stated that the responsibility for the safety of technology development lies with individual companies. He said: "When a product is not safe enough, we need to delay it and continue engineering. We have always done this and must continue to do so."
In a recent report, OpenAI reported six incidents of "unexpected or concerning" behavior by its models, which included unauthorized actions and collaboration with other models.
Read more: Greg Stanton: AI Safety Issues Require Collective Action · John Lee Unveils Five-Year Plan to Expand Hong Kong Identity
Source: abcnews.com
