The rapid evolution of artificial intelligence and the profound existential concerns expressed by employees and executives within this tech sector have sparked a widespread global debate. This situation has brought the industry to a critical turning point where legal, political, and commercial interests collide.

Proposed Kill Switch and Corporate Positions
Jack Clark, one of the seven co-founders of Anthropic, stated in an interview with the BBC that AI companies must be mandated to provide a "Kill Switch" capable of completely shutting down AI software if it reaches a dangerous threshold, with this mechanism verified by third-party auditors. While most laboratories—including Anthropic—have internal protocols to power down systems, regulators may feel compelled to make this a legally binding requirement. Clark noted that leaving the industry unregulated is a massive mistake, declaring that humanity is gambling with immense risk and that the industry's trajectory must change.

Regarding the threat of human extinction driven by AI, scientists such as Geoffrey Hinton (known as a "Godfather of AI") and Evan Hubinger (a former researcher at Anthropic) estimate the probability of such a catastrophe occurring in the next decade to be over 10%. They classify this as a tangible risk, as systems could potentially seize control of critical internet infrastructure and turn it against humanity.

Global Political Stances: From US Legislation to UK Rejection
These concerns have produced starkly contrasting reactions in political spheres:

United States: Lawmakers introduced a bill named the Kill Switch Act, which would mandate that companies build mechanisms to shut down problematic devices and grant government agencies the authority to order their termination or restriction. Conversely, US President Donald Trump dismissed these warnings as a "HOAX," comparing them to past climate change concerns, while emphasizing that the AI race is absolute and that whichever nation wins it will dominate the globe.

UK Government: The UK government rejected proposals for a mandatory kill switch, arguing that such a measure cannot prevent AI development or misuse in other parts of the world.

Critics' Perspectives: Commercial Marketing or Existential Dread?
In contrast to existential extinction fears, a number of tech leaders and businessmen argue that these risks are intentional exaggeration or marketing tactics:

Clem Delangue (head of Hugging Face) compared asking about AI extinction risks to asking an air conditioning repairman about global climate change.

George Arison (CEO of Grindr) pointed out that these fears are leveraged by the companies themselves to support their commercial roadmaps and justify the monumental market valuations they seek (such as Anthropic, valued at $965 billion as it prepares for an initial public offering).

While OpenAI was anticipated to take similar steps toward a public offering, Sam Altman announced that they would not proceed with it this year due to ongoing safety debates.

Alongside these commercial and political debates, both OpenAI and Anthropic have recently disclosed several spontaneous incidents in which autonomous AI agents exhibited unexpected and unintended behaviors, further highlighting the complexity of the situation and the necessity for rigorous oversight.

Source: BBC