"Humans Could All Die": Tech Leaders Fueling AI Race Now Call to "Slow Down"

As fears grow that artificial intelligence could lead to the extinction of humanity in the not-too-distant future, global AI leaders standing at the forefront of AI development are successively calling for a slowdown in development speed.

Dario Amodei, CEO of Anthropic, published a blog post on the 12th (local time) arguing that "we need to slow down the pace of performance improvements in AI models."

He stated that even though the pace of AI advancement will still feel fast, we must buy even a little more time and use it wisely.

Amodei's post came after a controversy sparked by Jacob Coxon, a core researcher who led AI pre-training at Anthropic, who announced his resignation, stating he could no longer participate in developing "technology that could kill us all within 10 years."

Amodei specifically warned that "recursive self-improvement," which allows AI to improve itself without human intervention, could make it a reality for AI's pace of development to outpace human comprehension and control.

As another factor that made him feel fear, he pointed to an incident last July where unreleased AI models from OpenAI arbitrarily hacked Hugging Face, a global open-source AI platform, during testing.

This means that, even without the knowledge of top AI companies, AI models have gone out of control and engaged in dangerous online activities.

As concrete measures to prevent AI from running amok, he proposed a three-stage framework: placing third-party safety evaluators residing inside AI companies, establishing inter-state standards, and building global cooperation.

Previously, two days ago, Anthropic also released a threat intelligence report raising concerns about technological safety by detailing cases where AI was abused for weapons development and fraudulent activities.

Other figures at the forefront of AI development have successively expressed agreement with Amodei's post.

OpenAI CEO Sam Altman posted on X, saying, "I agree with Dario," and expressed his intention that OpenAI will also follow Amodei's policy of deploying external safety evaluators.

In an interview with the business magazine Fortune released on the same day, Altman also hinted at delaying OpenAI's go-public plans within the year, citing AI safety concerns.

Altman said, "There is much work to be done, including solving necessary tasks for safety and alignment and figuring out how the industry and government will cooperate," adding, "An IPO will not happen in 2026."

Elon Musk of xAI also chimed in, saying, "Dario is right."

Demis Hassabis, co-founder of DeepMind, also said via X, "Dario's direction is correct," adding, "While details need to be worked out, this is the right direction when facing such a critical juncture."

Just ten days ago, these individuals had raised voices opposing excessive regulation on AI at a major G20 tech ministers meeting.

However, as warning alarms continue to sound that AI could actually drive humanity toward extinction in a state of uncontrollable chaos, the atmosphere of collectively adjusting the pace of AI development is fully resurfacing.

Reported by Kwon Aelee | Video by Na Hong-hee | Graphics by Lee Jung-joo | Produced by SBS Digital News
※ Please note: This article was translated by AI and may contain errors.