News

Legora and iManage plan to connect Legora to the iManage MCP Server, letting lawyers use governed iManage content and firm knowledge inside Legora.

Hogan Lovells Cadwalader will roll out Legora across its global network after a 1,300-lawyer pilot and act as a design partner for research tools.

Quantexa has joined the UN Global Compact, committing to its Ten Principles and to annual public reporting relevant to customers' vendor due diligence.

Stewart Tinson on who gives AI agents permission to spend, with this week’s coverage on board accountability, recovery and agent governance tools.

CoreWeave’s customer concentration and debt-funded expansion give bank boards a live case for combining credit, provider, market and operational AI risk.

Anthropic has given six METR staff employee-like access to assess its models and safety practices, with METR free to publish and Anthropic unable to veto.

The Bank of England’s bare-metal recovery focus moves AI cyber resilience beyond prevention and asks whether critical banking services can be rebuilt from trusted components.

The FSB has finalised 12 sound practices for responsible AI adoption by financial institutions, urging boards and senior managers to use them in governance.

The UK and Canada have renewed their AI security partnership, prioritising interoperable standards, Five Eyes information sharing and coordinated evaluations.

Treasury has announced more than $1.21bn in penalties against Bank of America over AML and sanctions failures, including unreported suspicious activity.

AvePoint has extended agent inventory and governance to Amazon and Claude agents and made Copilot Studio Agent Backup generally available in October updates.

IonQ has agreed to buy Zero Trust security company Elisity in an all-stock deal valued at up to $1.3bn, pairing it with IonQ's post-quantum cryptography.

Snyk's Evo Agentic AppSec uses AI agents that run Snyk's deterministic engines to find and fix vulnerabilities, verified independently of the agent.

Scale AI's new Scale Post-Training service customises open-weight models such as NVIDIA Nemotron 3.5 Super VL, already tested with enterprises Skydance and EY.

Pinecone BYOC now supports on-demand capacity, letting customers create indexes in their own AWS, Google Cloud or Azure account without sizing shards.