Aman Gupta

Anthropic says Project Glasswing partners using Mythos Preview found more than 10,000 high- or critical-severity flaws, raising pressure on fix validation, supplier exposure and containment

ServiceNow says a June update addressed an unauthenticated access issue, but customers still need tenant-level evidence of what was queried

After 5 million investigations, Google Cloud is expanding agentic AI across SOC triage, threat investigation and incident response, with humans retaining control.

Maricopa County's nearly 50-camera network shows how Arizona's AI wildfire detection buildout is moving from utility mitigation programs into county emergency-response operations

Unilever plans more than 40 new manufacturing digital twins with Accenture after reporting waste, capacity and quality gains at existing sites

New research argues that agent safety testing must follow tool calls, memory changes and system effects, not only the final response

The guidance details how to trace Copilot and Azure AI activity through Purview, Defender XDR and Sentinel, but reviewing prompt and response content still requires eDiscovery or DSPM for AI

Google Cloud says the UK security provider is using autonomous agents for 24/7 threat monitoring, making outsourced security operations a concrete SMB route for agentic AI

The IBV report says enterprises with stronger control across AI data, models and infrastructure protect 55% more operating profit from AI-driven disruption

Boston Children's, Harvard and OpenAI used AI to reanalyze 376 unsolved rare-disease cases, with clinicians confirming 18 diagnoses