# Goldman Sachs CEO David Solomon has revealed that the bank is in close collaboration with Anthropic following startling warnings regarding the cybersecurity threats posed by its latest AI models.

*business · news · 2026-04-14 · The News International*

## Key points

- Goldman Sachs is collaborating closely with Anthropic to address cybersecurity threats from Claude Mythos AI.
- Claude Mythos is the first AI to complete a 32-step cyber-attack simulation with a 30% success rate.
- Anthropic will not release Claude Mythos publicly, restricting access to select organizations under Project Glasswing.
- AISI warns that Mythos can autonomously target small and weak IT systems without human oversight.

**Companies:** Goldman Sachs, Google, OpenAI, Microsoft, Anthropic
**Countries:** United States, United Kingdom

[Read the full story on The News International](https://www.thenews.com.pk/latest/1398882-why-goldman-sachs-is-hyper-aware-of-anthropic-mythos-ai-risks)

---

Canonical: https://newsio.io/n/b6d09cbf-edec-4b87-91a9-eae140c8fd81/goldman-sachs-ceo-david-solomon-has-revealed-that-the-bank-is-in-close-collabora
Summarized by Newsio from The News International. https://newsio.io/how-it-works