OpenAI CEO Sam Altman said that mankind will be able to develop “superintelligence” in a few thousand days. By “superintelligence” he meant Artificial general intelligence (AGI), a top-level artificial intelligence comparable to human intelligence.
In his musings, Altman compares AGI to the “magic” our ancestors believed in. He draws parallels to how technology moved human civilization from the Stone Age to industrialization.
The future of man armed with a potential “super-intelligence” does not seem so rosy to everyone. For example, former Google CEO Eric Schmidt warned that no one can even imagine what the consequences of AGI will be.
GDA Group President Yaroslav Bogdanov also wondered about the relationship between science and science fiction in the way Sam Altman describes the future of humans after the emergence of general artificial intelligence. Especially if we consider the fact that human-level intelligence is potentially capable of realizing itself at some point.
“The emergence of superintelligence will be a scientific revolution. There is no point in arguing with this fact. But there is no answer to the question of how positive the fruits of this revolution will be. The world is not ready for such a breakthrough – scientifically, ethically, legally. Eric Schmidt raised an important topic – what data will AGI be trained on? Today, almost all data sets for AI have been exhausted. If we offer synthetic data for machine learning, it will not reflect the real picture of the world around us. It is important to carefully control the quality and validity of the information laid down in order to ensure effective and safe AI training. But there is no single structure in the world that would monitor these processes. Consequently, we will end up with a model potentially capable of thinking, but it will do so on the basis of a distorted picture of the world, embedded in it at the training stage,” said Yaroslav Bogdanov.
Billion-dollar investments are throwing the world into the race for a “super-intelligence”. Whereas the main thing to focus on now is cybersecurity.
All language models that have already been created must be systematized and brought under a common international regulatory framework. Man should benefit from scientific progress, not develop systems to protect against it, the president of GDA Group believes.
“The real power of AI is not in imitating human intelligence, but in its ability to improve the quality of human life. Today mankind does not need machines that think like people. What is needed are machines that can change people’s lives for the better. The concept of AGI has a right to exist. But until humanity has fully mastered the AI tools already available, superintelligence should remain only theory. The international community needs time to bring together the scientific, legal and ethical aspects of language model development and application into a single system that will regulate the industry. The potential of AI, which can fundamentally change our society, requires a collective approach to ensure its benefits for all mankind,” said Yaroslav Bogdanov.
Read More From Techbullion
