Goldman Sachs chief ‘hyper-aware’ of risks from Anthropic’s Mythos AI

Goldman Sachs CEO David Solomon says the bank is working 'closely' with Anthropic after warnings about the cybersecurity risks of its advanced AI model. | Contesto: cronaca

Punti chiave

• Goldman Sachs chief ‘hyper-aware’ of risks from Anthropic’s Mythos AI

Contesto

David Solomon, the chief executive of Goldman Sachs, has stated he is "hyper-aware" of the capabilities and risks associated with Anthropic's Mythos artificial intelligence model and is now engaged in close collaboration with the technology firm. The move comes directly in response to warnings issued by Anthropic itself about the potential cybersecurity threats its own advanced AI system could pose. The Wall Street giant had already been intensively monitoring rapid progress in AI, including the development of large language models (LLMs), as a core component of its broader strategy to defend against sophisticated cyber attacks. The bank's proactive stance underscores a critical and evolving dilemma at the highest levels of global finance: how to harness the transformative power of cutting-edge AI for competitive advantage while simultaneously fortifying defenses against the very same technology. Goldman Sachs is not merely a passive observer; it has integrated Anthropic's Claude model into its operations. This deep integration means the bank's exposure to the model's potential vulnerabilities is direct and operational, making the warnings from its creator impossible to ignore. Solomon's public acknowledgment of being "hyper-aware" signals a shift from theoretical risk assessment to active, urgent risk management. Anthropic's decision to proactively alert partners like Goldman Sachs to the dangers of its Mythos model represents a notable departure from the typical product rollout in the fiercely competitive AI industry. It suggests an internal assessment that the model's capabilities, potentially in areas like code generation, system analysis, or social engineering, could be repurposed by malicious actors to discover novel attack vectors, automate exploits, or bypass traditional security measures. For a financial institution like Goldman Sachs, a prime target for state-sponsored and criminal hackers, such capabilities present a clear and present danger to client assets, market stability, and proprietary trading systems. The collaboration between the bank and the AI lab is now focused squarely on improving cyber protection. This likely involves a multi-pronged...

Lettura DEO

Decisione di validazione: publish

Risk score: 0.1

Il testo è stato ricostruito dai dati editoriali disponibili senza aggiungere fatti non presenti nel record sorgente.

Indicatore di affidabilità

Verificata — Alta confidenza. Fonti affidabili confermano la notizia.

Il sistema a semaforo

Ogni articolo su DEO include un indicatore di affidabilità:

• 🟢 Verificata — Alta confidenza. Fonti affidabili confermano la notizia.
• 🟡 In evoluzione — Confidenza moderata. Alcuni dettagli potrebbero ancora cambiare.
• 🔴 Contestata — Bassa confidenza. Fonti in conflitto o incertezze rilevanti.

Questo sistema esiste perché chi legge merita di sapere non solo cosa è successo, ma anche quanto la notizia è solida.

Categoria: cronaca