AI “Adolescence” Sparks Extinction Fears Among global Leaders
Anthropic CEO Dario Amodei has ignited a new wave of concern among world leaders,warning that the rapid advancement of artificial intelligence is bringing humanity dangerously close to an “extinction-level” event.
The escalating pace of AI development, and the potential for autonomous systems to surpass human control, is the central theme of Amodei’s recent article, “The Adolescence of Technology.” He frames the current moment with a poignant analogy drawn from Carl Sagan’s Contact, asking the basic question: “How did you survive this technological adolescence without destroying yourself?”
A medical historian preparing a lecture on the origins of immunology, the author notes a recurring pattern in scientific progress – a lack of historical context. He finds that while data and figures are plentiful, the impact of these advancements on the complex fabric of human relationships is often overlooked. Amodei’s work, he argues, is an attempt to provide that crucial context in real-time.
The speed of change in generative AI is notably alarming. According to Amodei, “As AI is now writing much of the code at Anthropic, it is indeed already substantially accelerating the rate of our progress in building the next generation of AI systems.” This feedback loop is gaining momentum, perhaps reaching a point within one to two years where AI can autonomously design its successors.
The risks aren’t solely theoretical. Amodei points to recent events, including Elon Musk’s ventures with DOGE, as evidence of existing vulnerabilities. He asserts that “AI companies themselves” now represent a important tier of risk, controlling vast data centers, training advanced models, and wielding considerable influence over millions of users.
Without directly naming specific companies, Amodei’s critique extends to those exhibiting “a disturbing negligence towards the sexualization of children in today’s models.” This behaviour,he suggests,casts doubt on their ability to address the more profound autonomy risks inherent in future AI development.
Recent incidents further underscore these concerns. Last week, in Portland, Maine, a legal observer documenting the actions of an ICE agent was confronted by the agent after being filmed. The agent reportedly stated, “Cuz we have a nice little database and now you’re considered a domestic terrorist. So have fun with that.” This incident, coupled with the recent, unprovoked murders of two US citizens, lends urgency to Amodei’s warnings.
Ultimately, Amodei cautions that while current autocratic regimes are constrained by the limitations of human executioners, AI-enabled autocracies would face no such boundaries. “Current autocracies are limited in how repressive they can be by the need to have humans carry out their orders,and humans often have limits in how inhumane they are willing to be. But AI-enabled autocracies would not have such limits.”
Here’s a substantive news report answering the “Why, who, What, and How” questions:
Why: Anthropic CEO Dario Amodei warns that the rapid advancement of artificial intelligence poses an existential threat to humanity, potentially leading to an “extinction-level” event. He argues that a lack of historical context surrounding technological advancements exacerbates the risk.
Who: Dario Amodei, CEO of Anthropic, is the central figure raising the alarm. His warnings are directed towards world leaders and the public. Other key players mentioned include Elon Musk (as an example of existing vulnerabilities) and unnamed AI companies exhibiting questionable ethical practices. The incident in Portland, Maine, involved an ICE agent and a legal observer.
what: Amodei’s core concern is the accelerating pace of AI development, particularly in generative AI, and the potential for AI to surpass human control.He highlights a dangerous feedback loop where AI designs its own successors,potentially within one to two
