Top managers of companies that stand out in artificial intelligence technology made evaluations at the meeting held at the UNSC under the title "artificial intelligence and international security".
OpenAI Chief Executive Altman said that the speed in the development of artificial intelligence models is making the positive results more tangible, as well as "making the risks more urgent."
Altman said, "We have a choice before us. Artificial intelligence can either be like a new renaissance of creativity and discovery, or it can be like a new industrial revolution of shock and chaos." he said.
Stating that the ideal version of artificial intelligence is not "turning people into gears of a giant machine" or "optimizing every aspect of life until human elements disappear," Altman said, "These models can be new tools for people to realize their dreams and goals that seem unattainable today." he said.
Altman emphasized that artificial intelligence systems can become "more capable and autonomous", change the balance of power in the world, and eliminate human intervention as they improve themselves. "As the artificial intelligence development process becomes more and more automated, the pace of progress of artificial intelligence may increase significantly. This requires extreme caution." he said.
Noting that companies should not take risks for the sake of benefits and that no model should be trained if it is not sure that it can be kept under human control, Altman said, "For artificial intelligence to be democratic, the most important decisions should not be made only by laboratories in San Francisco. These decisions should be shaped through democratic processes and by governments that are accountable to their people." made his assessment.
Altman called for people to come together for common interests, saying, "We are at a crossroads." he said.
"I SEE TWO MAIN CATEGORIES OF RISK"
Anthropic's Chief Executive Amodei, in the online meeting he attended, stated that artificial intelligence has developed rapidly in recent years and explained that this extraordinary progress brings with it risks that need to be managed.
"I see two main categories of risk. The first is the misuse of AI models, for example, their use by bioterrorists to produce biological weapons. The second risk is loss of control; acceleration of model capabilities beyond the ability of developers to control," Amodei said. he said.
Stating that these risks are global issues that concern all humanity, Amodei said, "I believe that artificial intelligence may pose a risk to humanity as a whole if it is poorly managed." he said.
Pointing out that artificial intelligence security is a bigger issue than a single company can handle, Amodei said that managing risks requires collaboration across the industry and on a global scale.
Emphasizing that the work should start with narrow-scope agreements that every member of the Council can support, Amodei suggested banning the use of artificial intelligence in the production of biological weapons.
Amodei called for the establishment of evaluation and verification systems so that states can have information about the capabilities of the most advanced artificial intelligence models and verify whether the commitments made by countries to each other are fulfilled, and asked for common standards to be determined to test artificial intelligence models against risks.
Amodei said, "I believe this is the most important global security challenge facing the world today. Despite all the differences of the people in this room, we must put those differences aside to confront this global opportunity and global threat presented to us simultaneously. No leader, no company, and no nation can achieve this alone." made his assessment.
STRONGER STANDARDS IN ARTIFICIAL INTELLIGENCE GLOBALLY
Clement Delangue, Hugging Face CEO, explained the lessons they learned from this attack as the first company to publicly announce to the world the cyber attack carried out by an autonomous artificial intelligence agent.
Stating that stronger standards are needed on a global scale to better understand and reduce cyber security threats in artificial intelligence, Delangue emphasized that it has become a great need to distribute the capacity, resources and control of artificial intelligence instead of concentrating it in the hands of a few people.
Delangue stated that during the attack in question, they observed that artificial intelligence could cause fear in the public, especially through expressions that portray this technology as human.
Stating that they believe that artificial intelligence will be beneficial in terms of cyber security, despite being the victims of a cyber attack, Delangue said, "We were attacked by artificial intelligence, but more importantly, we defended ourselves using artificial intelligence. The same systems that helped us during this attack also help us in our fight against other cyber attacks that we are currently facing." he said.
