AI Governance

A new live experiment tests AI models’ ability to handle real management tasks under pressure, highlighting strengths and weaknesses in decision-making and trust.

A live experiment pits four top AI models against real business crises, revealing their decision-making personalities and whether they can be trusted to run a company.

SenseTime has issued guidance for the first half of 2026, but no specific figures or performance expectations have been disclosed, leaving the outlook unclear.

Exploring how established enterprises embed AI into their core systems, creating durable advantages and challenging assumptions about disruption.

Exploring how internal organizational feedback impacts the effectiveness of enterprise AI projects and why internal buy-in is crucial.

A report suggests Anthropic’s Claude may use a new watermarking method to identify AI-generated text, but technical details remain unconfirmed.

AI-driven talent density is transforming productivity, enabling small teams to outperform traditional organizations significantly. Key developments and implications explained.

Z.ai’s GLM-5.3 shows significant improvement in cybersecurity capabilities, raising new safety and governance questions amid its open-weight release.

Anthropic has launched a watermarking technique for its Claude AI system to aid content provenance, though technical details and effectiveness remain unclear.

Anthropic introduces machine-readable watermarks in Claude AI outputs, raising fears of restricted access in work and education.