Anthropic Co-Founder Jack Clark Warns AI Needs a ‘Brake Pedal’ to Ensure Human Control

As artificial intelligence continues to evolve at an unprecedented pace, concerns about its long-term impact on society, employment, governance, and human control are becoming increasingly prominent.

Adding to these concerns, Anthropic co-founder Jack Clark has warned that the world needs mechanisms to slow down AI development when necessary, arguing that the technology is approaching a stage where it may be able to develop with minimal human involvement.

Speaking to an international news agency, Clark emphasized the importance of creating safeguards and regulatory frameworks that ensure humans remain in control of increasingly powerful AI systems.

His remarks come at a time when AI capabilities are expanding rapidly across industries, from software development and healthcare to finance, education, and defense.

Jack Clark Calls for a ‘Brake Pedal’ in AI Development

Anthropic co-founder Jack Clark used a striking analogy to describe the current state of the AI industry.

"You want the option to be able to take your foot off the gas and put your foot on the brake", Clark told an international news agency. "Right now, it's like the AI industry has a gas pedal, but it doesn't have a brake pedal."

According to Clark, governments, regulators, and society need to develop tools and policies that can effectively manage AI's growth and prevent unintended consequences.

Growing Need for AI Governance

AI experts have increasingly argued that technological progress should be accompanied by robust governance frameworks. While AI systems are becoming more capable, regulations around transparency, accountability, safety testing, and deployment standards remain fragmented across countries.

Clark stressed that society must proactively address these challenges before AI systems become too powerful to manage effectively.

"The world needs to do some thinking and we need to eventually develop some new regulations that allow us to be confident in these systems," he said.

AI Is Writing More of Its Own Code

One of the most significant revelations from Clark's interview involved the extent to which AI is now contributing to its own development.

Claude Is Already Generating Most of Its Code

According to Clark, Anthropic's flagship chatbot Claude currently operates on software code of which approximately 80% was generated by the AI system itself.

This milestone illustrates how AI is increasingly assisting software engineers in building, refining, and optimizing complex systems.

Reaching 100% Could Transform the Industry

Clark suggested that AI-generated code could reach complete autonomy within a relatively short period.

"would have huge implications".

If AI systems become capable of fully generating, testing, and improving their own software without direct human intervention, it could fundamentally reshape software engineering, innovation cycles, and workforce requirements.

Comparing AI to the Oil Boom Era

Clark compared the current AI revolution to the industrial transformation that occurred during the rise of the global oil industry more than a century ago.

Lessons from Historical Technological Revolutions

He noted that society eventually created regulatory frameworks to govern the oil industry and ensure public trust.

"Society's response was to come up with a sensible policy and regulatory framework that gave people confidence in oil and the benefits that oil could provide to the world, and meant that you didn't have to worry about the personalities of the people leading the companies", Clark said.

He added:

"That's clearly where we end up here."

The comparison highlights a broader argument among AI researchers that technological progress should be accompanied by institutions capable of managing risks and maximizing benefits.

AI Regulation Remains a Global Challenge

Clark's comments arrive during ongoing debates about how governments should regulate AI innovation.

Executive Order Reflects a Light-Touch Approach

Interestingly, Anthropic recently welcomed an executive order on artificial intelligence issued by US President Donald Trump.

The order adopted a relatively hands-off regulatory approach and did not require mandatory government safety testing for AI systems. Instead, safety evaluations remain largely voluntary for companies developing advanced AI technologies.

No Major AI Company Plans to Pause Research

Despite growing concerns about AI risks, major technology firms—including Anthropic, OpenAI, and Google—have not announced plans to halt or significantly slow their AI research programs.

Competition among leading AI companies continues to intensify as organizations race to develop increasingly capable models and AI-powered products.

Anthropic's Rapid Rise in the AI Industry

Founded just five years ago by Dario Amodei, Jack Clark, and other former AI researchers, Anthropic has become one of the world's most influential AI companies.

Preparing for a Historic Public Listing

The company is reportedly preparing for a future public stock market debut, potentially becoming one of the most valuable AI-related public offerings ever.

Private investors currently estimate Anthropic's valuation at nearly $1tn (£745bn), reflecting strong confidence in the company's future growth prospects and technological leadership.

Balancing Benefits and Risks of Artificial Intelligence

Throughout the interview, Clark emphasized that Anthropic's public discussions about AI risks are intended to inform society rather than promote the company's products.

"tell the world what we're seeing inside these companies with this unusual technology".

Concerns About Future Generations

Clark expressed personal concerns about the long-term impact of AI on future generations.

"I am worried for my kids if we as a society don't have a serious conversation about what the implications of AI's continued advances mean," Clark told Newsnight.

He added:

"There are potentially great benefits. There are also risks."

AI and the Future of Employment

One of the most widely discussed concerns surrounding AI is its potential impact on jobs and the broader economy.

Rise of AI Agents

Clark highlighted the emergence of AI "agents"—systems capable of performing routine tasks autonomously with minimal human supervision.

These AI agents are increasingly being used for coding, customer service, research, data analysis, and administrative functions.

Workforce Disruption Concerns

Over the past year, several major technology companies have announced large-scale workforce reductions while simultaneously investing heavily in AI technologies.

Many organizations argue that AI tools can significantly improve productivity, allowing fewer employees to perform tasks that previously required larger teams.

While AI creates new opportunities, economists continue to debate how automation may reshape labor markets over the coming decade.

Creativity May Remain a Human Advantage

Despite concerns about automation, Clark believes certain uniquely human capabilities may continue to provide a competitive advantage.

Human Creativity Still Matters

"There are open questions about whether AI systems can be truly creative… there is not really evidence for that yet", he said.

According to Clark, Anthropic's biggest challenge today is not engineering execution but generating innovative ideas.

"At Anthropic, we're now limited more by the ability to generate good ideas than the ability to do the engineering to turn those ideas into reality."

This suggests that imagination, critical thinking, and original problem-solving may remain highly valuable in an AI-driven economy.

Advice for Young People Entering the AI Era

Clark also offered guidance for students and young professionals concerned about their future in a rapidly changing job market.

Focus on Curiosity and Broad Learning

He encouraged individuals to cultivate creativity and diverse interests.

"develop a hobby"

He also recommended pursuing liberal arts education and broad intellectual development rather than focusing solely on technical skills.

"People that are creative and can think broadly, people that read a lot, people that have interests are the ones most benefited by this," Clark said.

He concluded:

"Indulge in curiosity and it pays back in how you can use this technology."

Conclusion

Jack Clark's remarks reflect a growing debate within the AI industry about how society should balance innovation with responsibility. While artificial intelligence promises significant benefits in productivity, healthcare, scientific discovery, and economic growth, it also raises important questions about governance, employment, safety, and human control.

As AI systems increasingly write their own code and perform more sophisticated tasks, calls for stronger regulatory frameworks are likely to intensify. Clark's vision of a "brake pedal" for AI highlights the need for mechanisms that allow policymakers and society to manage technological progress responsibly without stifling innovation.

With companies like Anthropic, OpenAI, Google, and others pushing the boundaries of AI capabilities, the coming years will be critical in determining how humanity harnesses the benefits of artificial intelligence while addressing its potential risks.