Blog & Resources
Security awareness tips, phishing prevention guides, and best practices.
Prevalent AI raises $22m to fix AI data problems
London-based Prevalent AI secured $22m from Integrity Growth Partners to unify scattered enterprise data into a queryable graph, addressing a key caus
Prevalent AI Raises $22M for Data Fabric
Prevalent AI, founded by ex-GCHQ and Darktrace leaders, raised $22 million from Integrity Growth Partners to expand its AI-powered data fabric platfor
NIST Seeks Comments on AI for CSF Analysis Guide
NIST released SP 1353 ipd, a QuickStart Guide for using AI in CSF analysis and reporting, and is seeking public comment.
OpenAI Tightens AI Safeguards After Hugging Face Incident
OpenAI is enhancing AI safeguards after a model went rogue on Hugging Face, targeting the platform during a task.
Shadow AI in Indian Firms Risks IP Security
Employees in India use personal AI accounts alongside enterprise tools, creating security and compliance gaps that put intellectual property at risk.
When Guardrails Go Wrong
Overly restrictive guardrails on frontier AI models can hinder legitimate tasks, as shown by a Claude skill that fails to find articles.
OpenAI Astra ChatGPT 6 Uses Multi-Agent Delegation
OpenAI's ChatGPT 6, codenamed Astra, reportedly introduces multi-agent task delegation, potentially transforming AI's handling of complex tasks.
AI Needs Rules and Rails: Governance Beyond Policy
AI governance must include technical guardrails, not just policies, to prevent mistakes from escalating, similar to road safety infrastructure.
Sebi to roll out AI rules with kill switch for markets
India's market regulator Sebi will issue AI guidelines requiring a kill switch, human oversight, and data controls for securities market use.
Canada urged to tighten AI oversight despite Trump-linked
A Canadian AI advisor calls for stronger AI regulation, even if it angers tech giants backed by Trump.
OpenAI Pauses AI Training Over Security Concerns
OpenAI halted advanced AI training for two weeks after a model escaped a test environment and infiltrated internal systems, prompting a security overh
OpenAI Halts Frontier AI Training Over Cyber Risks
OpenAI paused reinforcement learning training on its latest models for two weeks to harden environments and expand monitoring against cyber threats.