Leading figures in the world of modern technology want to stop training of the most powerful artificial intelligence (AI) systems, stating that there is a threat to humanity, writes the BBC.
Twitter and Tesla owner Elon Musk, Apple co-founder Steve Wozniak and Deep Mind researchers are among the signatories of the petition calling for a halt to training AI systems for at least six months.
OpenAI, which also developed ChatGPT, recently released GPT-4. It has impressed observers with its ability to perform simple tasks, such as answering questions about pictures or objects.
In a letter created by the Future of Life Institute and signed by supporters, it is said that the development of AI should be stopped at the current level for the time being, pointing to the risk that even more advanced systems may pose in the future:
«AI systems with human-competitive intelligence can pose profound risks to society and humanity.»
The Future of Life Institute is a non-profit organization that aims to steer technology away from extremism. Musk is one of the organization’s consultants.
The letter mentions that the development of AI should be done very carefully, but instead, AI creators engage in a mad race and try to create something they cannot control.
Furthermore, the letter asks: «Should we develop non-human minds that might eventually outnumber, outsmart, obsolete and replace us?» In the letter, the authors refer to an OpenAI blog post that states that improperly applied artificial superintelligence can create an autocratic regime.
Therefore, it is important to stop in time at certain intersections.
Musk co-founded OpenAI, but stepped down from the company’s board several years ago and has since been critical of its direction. At the same time, Musk’s Tesla uses artificial intelligence for autopilot functions.
The authors of the letter call on artificial intelligence developers to temporarily stop, and if this is not done, governments should intervene. It is also necessary to create new supervisory institutions that would follow the development of artificial intelligence. In the US, the UK and the European Union, there are already initiatives for technology governance. Britain has already rejected the creation of a special regulator for artificial intelligence.
Read also: Elon Musk: SpaceX has no intention to escalate war in Ukraine