AI reveals what the biggest thing humans have to be scared of when it comes to artificial intelligence
ChatGPT is warning people about the 'slow creep of irreversible dependence on systems we don’t fully understand'
An AI bot has revealed what the biggest thing humans have to be scared of is when it comes to artificial intelligence.
The artificially intelligent chatbot, ChatGPT, has warned what it is that humanity has to be scared of when it comes to the technology.
When posed with the question, the bot explained that ‘once AI starts running key parts of society—like defense, finance, infrastructure, healthcare, and governance—we may not be able to stop it, understand it, or undo its decisions’.
This will result in a loss of human control over critical systems.
Advert
So, why is this terrifying? Well ChatGPT has an answer.
It explained: “Advanced AI systems can make decisions in microseconds. Humans just can’t keep up. By the time we notice something’s wrong, it could be too late.
“Many AI systems (especially deep learning models) are ‘black boxes’. We don’t know exactly how they reach their conclusions. So if an AI decides to launch a preemptive strike or crash a stock market... how do we question it?
“Imagine two AIs facing off in a military standoff. One interprets a move as hostile, retaliates instantly, and the other responds in kind. Humans aren’t even in the loop. Boom—escalation spiral.
“The more we rely on AI, the more helpless we become without it. Eventually, we won’t choose to give AI control—it’ll feel like we have to, just to stay competitive or keep things functioning.”
And ultimately, the tech doesn’t even have to be a villainous power-hungry entity, like we often see in the movies, to do a lot of harm to the world.
ChatGPT continued: “Even if AI isn't 'evil', it can still cause catastrophic harm just by not understanding human values correctly. Example: You tell an AI to ‘stop climate change’, and it decides humans are the biggest polluters…”
But what does this mean in real terms? Well, it’s possible that AI could play a role in nuclear command-and-control systems, controlling financial markets, autonomous drones making kill decisions in real-time, manipulating public opinion at scale, and AGI (Artificial General Intelligence) pursuing goals misaligned with human well-being.
The bot continued: “It’s not the Terminator scenario that should keep you up at night—it’s the slow creep of irreversible dependence on systems we don’t fully understand, that operate faster than we can manage, and that might one day treat humans as an inefficiency to optimize.”
And coming from an AI bot, that’s pretty scary stuff!