AI Employees Skeptical of Doomsday Scenarios
Employees at major tech companies are skeptical of claims that unregulated artificial intelligence will bring about the end of humanity.
Employees at leading artificial intelligence companies such as OpenAI, Meta, and DeepMind are maintaining a distance from claims that the uncontrolled development of the technology will lead to the extinction of humanity.
Employees' Skeptical Approach
Some employees working at companies like OpenAI, Meta, and DeepMind are skeptical of the idea that uncontrolled artificial intelligence will lead to mass deaths. It is stated that there is a lack of sufficient detail behind existential fears on this subject.
Diverse Voices in the Industry
Claims by former Anthropic employee Jacob Coxon had resonated in the industry, leading to calls to slow down development. However, Meta data scientist Colin Fraser stated that there is no real evidence that artificial intelligence will pursue a goal that would cause human death.
Near-Term Real Risks
Although employees joke about existential scenarios, they share common concerns regarding immediate risks such as bypassable firewalls and military use. This situation became even more pronounced with the loss of control experienced during safety tests.
Search for Independent Evaluation
There is a growing consensus regarding placing independent evaluators from AI safety research organizations into major AI laboratories. Anthropic announced that it will bring in AI evaluators from Faculty.