Anthropic’s Safety Superpower – Stratechery by Ben Thompson

kept by eddie jun 16

Anthropic's genuine belief in safety licenses it to prioritize business interests and challenge the U.S. government.

Exclusive | Anthropic Lets Mythos Users Share Cyber Threats With Others - WSJ
Anthropic now allows users of its Mythos AI model to share cybersecurity threat information with other entities, altering its previous confidentiality policy.
also on Mythos, #ai-safety, #enterprise, #us-government, Anthropic, #ai
Anthropic Claims Its New A.I. Model, Mythos, Is a Cybersecurity ‘Reckoning’ - The New York Times
Anthropic says its unreleased AI model Mythos can prevent cyberattacks and is partnering with 40 companies to test it.
also on Mythos, #ai-safety, #enterprise, Anthropic, #ai
You're Spending Too Much on AI. You're Also Using Too Little.
Companies overspend on AI by using costly frontier models for routine work, needing better defaults, outcome-based measurement, and cultural efficiency incentives to unlock far greater value.
also on Fable, OpenAI, Anthropic, #ai
Prompt Caching In Agents
Prompt caching is crucial for coding agents’ cost and latency, but fragile; Pi maintains stable inputs and exposes cache metrics to help users manage performance.
also on Fable, OpenAI, Anthropic