**
In a striking call for caution, Jack Clark, co-founder of Anthropic, has urged for a regulatory ‘brake pedal’ on artificial intelligence (AI), warning that the technology is advancing toward a point where it could evolve independently of human oversight. Speaking on BBC Newsnight, Clark emphasised the need for policymakers to intervene and establish a framework that ensures human control over increasingly powerful AI systems.
The Need for Regulatory Control
“You want the option to be able to take your foot off the gas and put your foot on the brake,” Clark remarked, highlighting the current lack of mechanisms to slow down AI development. He asserted that while the AI sector has accelerated its advancements—evident in Anthropic’s chatbot Claude, which can already generate 80% of its own code—there must be a concerted effort to create regulations that instil confidence in these technologies.
Clark’s comments come amid growing concerns that AI could outpace human governance, with potential implications that extend beyond technology into the fabric of society. “The world needs to do some thinking,” he stated, calling for new regulations to manage the societal impacts of AI.
The Path to 100% Autonomous Code
Anthropic’s chatbot, Claude, is already showcasing remarkable capabilities, with Clark predicting it could soon reach the point of writing all its own code. “Getting to 100% is possible within two years,” he noted, warning that such advancements could have significant ramifications. Without proper oversight, the risk of AI systems operating beyond human control becomes increasingly likely.
While Clark did not detail specific strategies for implementing regulatory measures, he drew an analogy with the early days of the oil industry, where societal responses led to sensible policies that ensured safe practices. “That’s clearly where we end up here,” he explained, underscoring the importance of establishing trust in AI technologies.
Anthropic’s Rapid Growth and Market Position
Despite advocating for caution, Anthropic is poised for a significant market debut, preparing for one of the first public listings among newer AI firms. Valued at nearly $1 trillion (£745 billion) by private investors, the company stands to make a historic entry into the stock market. Clark emphasised that their public discourse on AI capabilities is not merely a tactic to attract clients but a sincere effort to share insights into the evolving landscape of AI technology.
Founded by Clark and a group of former OpenAI employees, Anthropic has carved a reputation for addressing the potential hazards of AI. The company has even engaged in public disputes over the ethical implications of their technology, particularly concerning its potential applications in mass surveillance and autonomous warfare.
The Societal Implications of AI Advancement
As AI continues to evolve, Clark expressed concern over its economic implications, particularly regarding job displacement. The rise of AI-powered agents capable of performing tasks autonomously has already led to significant layoffs in major tech firms, as the industry grapples with the reality of machines replacing human roles.
However, Clark remains optimistic about the future for creative individuals. He believes that those with innovative ideas will hold an advantage in this AI-driven landscape. “There are open questions about whether AI systems can be truly creative… there is not really evidence for that yet,” he stated, encouraging young people to pursue diverse interests and liberal arts education as a means to thrive amid AI advancements.
Why it Matters
Clark’s urgent call for AI regulation signals a pivotal moment in the ongoing discourse surrounding technology and governance. As AI capabilities grow, the imperative for a structured approach to its development becomes more pressing. Striking a balance between innovation and safety will not only protect jobs but also ensure that the benefits of AI are harnessed responsibly, ultimately shaping a future where technology serves humanity rather than replacing it.
