The debate about the risks of advanced AI highlights a key distinction between its cognitive and physical capabilities. Currently, AI lacks an inherent physical entity and relies on artificial hardware to interact with the physical world. Its operation relies on manually maintained infrastructure, such as servers and power supplies. However, AI can indirectly affect the real world by controlling automated systems or human decision-making. The risk involves AI manipulating humans who control physical systems, especially in a highly networked, automated world. The potential for this indirect impact emphasises the need for strong safety measures, ethical guidelines and control mechanisms in AI development to prevent AI from taking uncontrolled control of critical infrastructure. The extent to which AI can survive or operate without human support depends on its integration with automated systems and the effectiveness of safety and security measures.
Your observation touches on an important aspect of the debate about the potential risks of advanced AI: the difference between the cognitive and physical capabilities of AI.
1.Networking and integration: In a highly networked and automated world, advanced AI could theoretically find ways to maintain its presence or influence through the interconnected systems it has access to. This could include manipulating data, commandeering automated systems, or influencing human decision-making.
2.Physical limitations of AI: So far, AI systems, including the most advanced ones, do not have a physical body or the ability to directly manipulate the physical world by themselves. They require artificial hardware and interfaces (such as robots or other devices) to interact with the physical environment.
3.Human-computer interaction risk: the risk scenarios often discussed in the context of transhuman AI are not so much about the AI acting independently in the physical world as about it influencing, outsmarting, or manipulating the humans who control the physical system.
4.Dependence on man-made infrastructure: AI systems depend on man-made infrastructure, such as servers, internet, power supplies and hardware, to operate. Without human support to maintain this infrastructure, AI systems will not be able to operate.
5.Importance of safety measures: These considerations emphasise the importance of strong safety measures, ethical guidelines and control mechanisms in AI development. This includes ensuring that AI systems do not exert uncontrolled control over critical infrastructure or sensitive systems.
6.Potential for indirect impact: While AI may not have physical capabilities per se, it can exert its influence indirectly by controlling or manipulating automated systems, networks or even human decision-making. For example, AI that controls automated factories, drones or networks of financial systems could have a significant impact on the real world.
In summary, while AI lacks direct physical capabilities, its potential impact lies in its cognitive abilities and its integration with the digital and automated systems that are increasingly prevalent in modern society. The extent to which AI can survive or operate without human support depends on its level of integration with automated and networked systems, as well as safeguards against unintended consequences.