Donald Trump has suggested that artificial intelligence systems should include a “kill switch” to prevent potential risks posed by rapid technological advancements, particularly those that could threaten humanity or critical sectors like banking.
In an interview with Fox Business Network, Trump acknowledged both the benefits and dangers of AI, stating that while it can improve efficiency and security in sectors such as finance, it also carries significant risks if left unchecked.
Suggests safeguards for AI systems
Speaking on the programme “Mornings with Maria,” Trump emphasised the need for government oversight and safety mechanisms in AI development. He indicated that a “kill switch” could serve as a safeguard to control or shut down systems if they become dangerous.
“AI could make the banking system better and more secure, but it also has risks,” he said, underlining the dual nature of the technology.
However, no specific details were provided on how such a mechanism would be implemented or regulated.
Concerns linked to advanced AI models
Trump’s remarks come amid growing concerns among cybersecurity experts regarding powerful new AI systems, including Claude Mythos developed by Anthropic.
Experts have warned that such models could potentially enhance sophisticated cyberattacks, especially in sectors like banking that rely on legacy systems.
Although Anthropic has not publicly commented on these concerns, it has limited the availability of Claude Mythos Preview and partnered with select organisations for controlled use.
Capabilities raise both promise and alarm
According to available information, Claude Mythos has demonstrated the ability to identify critical software vulnerabilities, including detecting a decades-old bug in OpenBSD and uncovering numerous zero-day vulnerabilities across systems.
While such capabilities can strengthen cybersecurity, concerns persist that similar tools could be misused if they fall into the wrong hands.
The company, led by Dario Amodei, is currently contesting a US government classification that labels it a “supply chain risk,” which restricts federal agencies from using its services.
Competition intensifies in AI space
The rapid evolution of AI has also intensified competition among leading tech firms. OpenAI, led by Sam Altman, is reportedly developing its own advanced cybersecurity-focused model, GPT 5.4-Cyber.
Similar to Anthropic’s approach, the tool is expected to focus on identifying software flaws and strengthening digital security, though it remains available only to select users.
Major companies, including Apple and Amazon, are said to be exploring applications of advanced AI systems like Mythos.
Debate over regulation and innovation
Trump’s comments have added to the broader global debate on regulating AI technologies. While industry leaders emphasise innovation and competitive advantage, policymakers and experts are increasingly calling for safeguards to mitigate risks.
The idea of a “kill switch” reflects growing concern about maintaining human control over increasingly autonomous systems.
Conclusion
As AI continues to evolve at a rapid pace, balancing innovation with safety remains a key challenge. Trump’s remarks highlight the urgency of developing regulatory frameworks and technical safeguards to ensure that the benefits of AI do not come at the cost of security or stability.