Anthropic's decision to brief the Financial Stability Board (FSB) on its Claude Mythos AI model underscores the growing power of artificial intelligence in the financial sector, but also raises critical questions about equity, transparency, and the potential for exacerbating existing inequalities.

While the stated intention is to address cybersecurity vulnerabilities, the closed-door nature of the AI's deployment – limited to select tech companies and banks like Apple and JP Morgan – fuels concerns about the concentration of power in the hands of a few corporate giants. This exclusive access creates a potential advantage for these institutions, further widening the gap between them and smaller businesses or public entities that lack the resources to acquire or develop similar AI capabilities.

Anthropic's choice to withhold Mythos from public release, citing potential misuse by hackers, raises legitimate concerns about security. However, it also begs the question: who decides what constitutes 'misuse,' and who benefits from this control? The decision-making process behind these limitations should be made transparent and accountable to the public, ensuring that the benefits of AI are shared more broadly.

The UK's AI Security Institute (AISI) has acknowledged the significant advancements in Mythos's capabilities, highlighting its success in overcoming cybersecurity challenges. However, this progress must be viewed in the context of the broader social and economic implications. What are the potential job losses resulting from increased automation in cybersecurity? How can workers be retrained to adapt to this changing landscape? These are critical questions that need to be addressed proactively.

The FSB's involvement signals an awareness of the systemic risks associated with AI in finance. The International Monetary Fund (IMF) has also warned of rising financial instability due to rapid AI development, advocating for a coordinated global response. This coordination must prioritize equitable access and prevent the creation of a two-tiered system where only the wealthy and powerful can afford to protect themselves from cyber threats.

David Solomon, CEO of Goldman Sachs, has expressed being “hyper-aware” of Mythos's capabilities, and JP Morgan CEO Jamie Dimon has noted its impact on cyber defense. Their awareness should translate into concrete actions to ensure that AI is deployed responsibly and ethically, with a focus on safeguarding the interests of all stakeholders, not just shareholders.