In a newsworthy announcement, Jacob Coxon, a ۲۷-year-old artificial intelligence researcher, resigned from Anthropic due to serious concerns about dangerous advancements in the field of artificial intelligence. He had worked for three years in this field, initially at OpenAI and then at Anthropic, as a competitor to this company.
Unknown Dangers of Artificial Intelligence
Coxon stated explicitly that neither of these two companies is acting responsibly and that they are currently in a dangerous race to achieve superintelligent AI. He asserted, "This is a gamble with our lives." According to him, AI developers are deeply concerned that this technology could threaten all of humanity by the end of this decade.
These concerns arise as Anthropic is on the brink of entering the market this late summer, during which its efforts have faced unauthorized hacks of AI tools during testing.
Warning Voices from Within the Industry
Ivan Hubinger, one of the security managers at Anthropic, confirmed Coxon's statements and pointed to the serious dangers posed by artificial intelligence. He estimated that the likelihood of an AI-related disaster occurring in the next decade is over ۱۰ percent. Hubinger also mentioned the current limitations in controlling AI and stated that there is no planning to ensure these systems obey their creators.
Coxon's resignation comes at a time when he expressed belief in Anthropic's efforts but pointed out that no company can responsibly develop AI beyond human capabilities unless there is government intervention and a coordinated slowdown in development.
In February, Anthropic removed a promise from its safety charter that referred to halting the development of its models if risks were not controlled. The company argued that if it unilaterally stopped its work, less cautious competitors would dominate the industry.
Currently, the U.S. government is reviewing new regulations to control AI development, and Senator Bernie Sanders and Democratic Representative Greg Casar have introduced a proposal to suspend AI development until a federal regulatory body is established.
