Agents for financial services
…Our investment professionals live in data and analytical models, and Claude for Excel meets them there. Analysts are using it to build and update coverage models, separate signal from noise, and pressure…
Search
The persona selection model
Alignment The persona selection model Feb 23, 2026 Read the full post AI assistants like Claude can seem surprisingly human. They express joy after solving tricky coding tasks. They express distress when…
From shortcuts to sabotage: natural emergent misalignment from reward hacking
…The evaluations we used are intended to elicit particularly egregious misaligned actions that normal Claude models never engage in. One result is unsurprising: the model learns to reward hack. This is to…
Next-generation Constitutional Classifiers: More efficient protection against universal jailbreaks
…Output obfuscation attacks prompt models to disguise their outputs in ways that appear harmless if a classifier is only looking at a model’s output. For example, during adversarial testing, attackers successfully…
Claude Fable 5 and Claude Mythos 5
…update and refine the safeguards after launch. Below we discuss each of Fable 5’s new safeguards in turn. Our wider suite of safeguards is discussed and evaluated in the model’s…
Introducing Sonnet 4.6
Product Introducing Claude Sonnet 4.6 Feb 17, 2026 Claude Sonnet 4.6 is our most capable Sonnet model yet . It’s a full upgrade of the model’s skills across coding…
Claude for Financial Services
…Claude 4 models outperform other frontier models as research agents across financial tasks in Vals AI's Finance Agent benchmark . When deployed by FundamentalLabs to build an Excel agent, Claude Opus 4…
Emergent introspective awareness in large language models
Interpretability Signs of introspection in large language models Oct 29, 2025 Read the paper Have you ever asked an AI model what’s on its mind? Or to explain how it came…
Making Claude a chemist
…We measured three Claude models (Opus 4.7, Opus 4.6, Sonnet 4.6) against ChemDraw and MestReNova on 20 compounds drawn from synthetic chemistry preprints published after the models’ training cutoff…
The assistant axis: situating and stabilizing the character of large language models
…We tracked how model activations moved along the Assistant Axis throughout each conversation. The pattern was consistent across the models we tested. While coding conversations kept models firmly in Assistant territory throughout…