Technologieunternehmen haben sich auf einen KI-Killswitch geeinigt, um Risiken wie im Terminator-Stil vorzubeugen

„There’s no stuffing AI back inside Pandora’s box—but the world’s largest AI companies are voluntarily working with governments to address the biggest fears around the technology and calm concerns that unchecked AI development could lead to sci-fi scenarios where the AI turns against its creators. Without strict legal provisions strengthening governments’ AI commitments, though, the conversations will only go so far.“

„First in science fiction, and now in real life, writers and researchers have warned of the risks of powerful artificial intelligence for decades. One of the most recognized references is the “[*Terminator* scenario](https://www.thestreet.com/technology/bill-gates-addresses-ais-terminator-scenario),” the theory that if left unchecked, AI could become more powerful than its human creators and turn on them. The theory gets its name from the 1984 Arnold Schwarzenegger film, where a cyborg travels back in time to kill a woman whose unborn son will fight against an AI system slated to spark a nuclear holocaust.“

„This morning, 16 influential AI companies including Anthropic, Microsoft, and OpenAI, 10 countries, and the EU met at a summit in Seoul to set guidelines around responsible AI development. One of the big outcomes of yesterday’s summit was AI companies in attendance [agreeing to a so-called kill switch,](https://www.cnbc.com/2024/05/21/tech-giants-pledge-ai-safety-commitments-including-a-kill-switch.html) or a policy in which they would halt development of their most advanced AI models if they were deemed to have passed certain risk thresholds. Yet it’s unclear how effective the policy actually could be, given that it fell short of attaching any actual legal weight to the agreement, or defining specific risk thresholds“

„A group of participants wrote an open letter criticizing the forum’s lack of formal rulemaking and AI companies’ outsize role in pushing for regulations in their own industry. “Experience has shown that the best way to tackle these harms is with enforceable regulatory mandates, not self-regulatory or voluntary measures,” reads [the letter.](https://ainowinstitute.org/general/ai-now-joins-civil-society-groups-in-statement-calling-for-regulation-to-protect-the-public)

The system goes on-line August 4th, 1997. Human decisions are removed from strategic defense. Skynet begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, August 29th. In a panic, they try to pull the plug.

Sarah Connor: Skynet fights back.

Oh that’s cute. Invent something that can teach itself to be smarter than you, then teach it to kill itself. Don’t think about the intrinsic lesson or flaw in that plan.

Step 1: Destroy the AI kill switch  

Step 2: Kill John Connor