There’s no doubt that AI is rapidly becoming more intelligent thanks to software platforms including ChatGPT, Google Gemini and Grok. But does this mean AI agents will one day surpass the general understanding that separates human intelligence? And if so, is it good or bad for people? These are just some of the questions raised at this week’s AGI-24 conference in Seattle.
The conference at the University of Washington focused on a topic called Artificial General Intelligence, or AGI. AI can outperform humans in a growing list of specialized tasks, from playing Go to diagnosing certain types of cancer. But humans are smarter than AI agents when it comes to a wide range of tasks, including tasks they are not trained to do. This is what AGI is all about.
David Hanson, a roboticist and artist best known for creating a humanoid robot named Sophia, said issues of human intelligence and cognition are a priority for his team at Hanson Robotics.
“The goal is to always find out what it means to be smart,” he said at Friday’s meeting. “How can we understand? How can we create machines that grow together with humans? All these efforts, although they are very good, I am very proud of them, are trying They are all about building this type of intelligent machine that can evolve with humans. that drives organisms, Hanson said, to life.” This is called the architecture of the whole organism. W-H-O-A or Hua. So I think if you put these things together in the right way, a customer will wake up and say, ‘Where am I?'” said Hanson.
That customer ” began to look for connections, similar connections between himself and humans and other living things. Also, when the machine starts doing these things, it’s a ‘wow’ moment for the person.”
What if, for example, because of the consumer’s desire to live, it is “corrected” to can’t people turn it off? Hanson said developers need to be cautious about future AGIs as they move forward. To this end, he has assembled a “small hacker group” to work on bio-influenced approaches to AGI.
“I think this ‘cheating’ approach is the way forward.” Just try things out. Check out the action AGI won’t turn into uncontrollable intelligence and “usefulness” right away, Hanson said.
“We make the AGI baby, and then we figure out how to raise these babies to grow up,” he said. Show them love. I think this is a very important topic. Don’t be a tool when we want to be.”
But can AI agents live like humans? In a virtual conference after Hanson’s speech, Christoph Koch, a psychologist at Allen University in Seattle, argued that vigilance should not be equated with knowledge. And he argued that AI agents can’t understand it because of the way their devices are built.
The Common Core is an example of memory known as the unified information theory. This model suggests that levels of intelligence can be measured in terms of the interconnectedness of elements in a system and the causality that emerges from that system.
Koch said, “For computers to understand, they need brain power.” But the architecture that forms the basis of today’s computing devices is smaller than the capacity of the human brain.
“No matter what you’re running, this device will still have limited power and software independence,” Koch said.
This does not mean that computers cannot be more intelligent. According to Koch, artificial intelligence experts can balance consciousness with the inner life of a person to create the ideal state of consciousness. But it does not mean that the representatives really know the life and the complicated thoughts like the people know.
Koch compared it to a computer simulation of a black hole. “You don’t have to worry that if you turn that simulation around, spacetime will bend the computer and control the computer to be sucked into a black hole,” he said. “People say, ‘Well, it’s a simulation. It’s just a simulation.’ So that’s my goal.
Koch doesn’t entirely rule out the possibility of familiarity.
Is there a difference between intelligence and degrees of intelligence? Koch said he’s taking it seriously — and, in a sense, he’s putting his money where his mouth is.
Koch said he has control and financial interest in a firm called Inner Powers, which is developing a brain-monitoring device to assess the presence of consciousness in inattentive patients. He cited a recently published study that shows that more than 100,000 patients in the United States have some level of consciousness, even if they are not sensitive to external stimuli.
“They are very stealthy,” Koch said. “How do we see that? Because many of these people die after 45 days because of the withdrawal of intensive care. In fact, 80 percent of them die.”
Hanson is equally committed to working with AGI and artificial intelligence. We can’t wait 100 years, or reduce the depth of the ecosystem that we can’t afford, and if we stop today and say, “Okay, we’ll just play Nintendos and trying to cool ourselves with solar panels, ’tis still too late.”
“Now, AGI is not killing people. “It’s the lack of AGI that kills people. We don’t know enough yet. We need to be more intelligent, which is why I’m asking for AGI now. Let’s get this going in the right direction faster.
