收听本期播客
阅读正文
British banks are poised to gain access to Mythos, a sophisticated new artificial intelligence tool developed by Anthropic. This development is generating considerable debate, as Mythos was previously deemed too risky for widespread public release, sparking concerns among global financial leaders.
Anthropic, already known for its Claude series of AI systems, had until now limited Mythos to a select group of major US companies, including giants like Amazon and Microsoft. However, Anthropic’s UK operations head, Pip White, recently confirmed that access would soon extend to financial institutions in the UK, potentially within the next week. This decision is noteworthy, given the company’s own earlier warnings about the tool’s advanced capabilities.
Mythos presents a unique challenge due to its exceptional ability to identify and exploit vulnerabilities in IT systems. Anthropic itself has acknowledged that current AI models possess coding skills that surpass most human experts in discovering software weaknesses. If this technology were to fall into the wrong hands, the potential impact on national security, public safety, and global economies could be extremely severe.
These significant threats have been a central topic at recent international gatherings of finance ministers, executives, and regulators, including the IMF and World Bank spring meetings. Canadian Finance Minister François-Philippe Champagne described the challenge as an “unknown unknown,” emphasising that the full scope of the risks is not yet understood. He called for urgent attention to establish safeguards for the financial system’s resilience.
Andrew Bailey, the Governor of the Bank of England, recognised the seriousness and rapid pace of AI advancements. He posed a crucial question for regulators: what is the “optimum moment” to set rules for such technology? Acting too soon might hinder innovation or miss the mark, while delaying could allow situations to become uncontrollable. Christine Lagarde, President of the European Central Bank, echoed this sentiment, highlighting the pressing need for a robust governance framework to manage these powerful tools responsibly.
US Treasury officials have already engaged with major American banks concerning Mythos, focusing on institutions whose disruption could jeopardise financial stability. Similar discussions are anticipated between UK regulators, bank executives, and government officials in the coming weeks. The global financial community is clearly striving to balance the advantages of advanced AI with the necessity of effectively managing its considerable risks.