GPT-5 may behave like a human and be released this year
ChatGPT developer Siqi Chen dropped a bombshell regarding its development. He tweeted that GPT-5 may launch in December.
The term AGI refers to artificial general intelligence, which is capable of thinking and learning in a similar way to humans. In other words, it would be virtually impossible to tell ChatGPT from a real person.
As we’ve seen with the AI chatbot lately, it has been able to do amazing things, so it’s easy to believe such a claim. However, OpenAI hasn’t provided any further details as of yet.
A member of Sandbox VR’s board of directors and co-founder of Runway Financial, Siqi Chen commented on GPT-5:
“I have been told that GPT-5 is scheduled to complete training this December and that OpenAI expects it to achieve AGI,” he said.
“Which means we will all hotly debate as to whether it actually achieves AGI,” he continued.
“Which means it will,” Chen concluded. According to his prediction, we will have an AI that learns and thinks like humans this year. Other experts expected it to arrive in 2032.
In the real world, we have never seen artificial intelligence that thinks and learns like us. Only science fiction has done this.
Nevertheless, let’s speculate on its potential benefits. First, it will be able to perform tasks with human-like accuracy. With AGI, ChatGPT would no longer require detailed information when performing tasks. It could instead deduce and infer solutions just like a person.
A second benefit of GPT-5 is that this human-like reasoning will be paired with massive computational power, allowing us to solve issues that could not be solved in the past.
For example, it can defend someone in court because it understands human speech. Additionally, it has access to hundreds of pages of criminal law to find a defense.
A digital environment was used to simulate numerous COVID-19 tests. That took months, but GPT-5 may provide solutions in weeks!
The ChatGPT could be a real companion for people if it can think like a human. Tesla has been working on making a robot like that, and the ChatGPT might be the key to achieving that.
The problem of providing rights to artificial intelligence, however, might prove to be difficult. After all, it would demand that we humans treat it equally. Multiple factors could lead to artificial general intelligence harming humans. Worse yet, it could gain access to the internet.
Since the world is connected to the internet, GPT-5 may become increasingly powerful.
