A founding member of OpenAI has joined Anthropic to boost Claude’s research capabilities

OpenAI co-founder Andrej Karpathy is now part of Anthropic’s AI team and will helm a team that handles pre-training research.
…

Microsoft confirms two major Defender security issues — so update now or face possible attack

CISA confirms two bugs being actively exploited in the wild, as Microsoft releases patches.

“£400 for a fan? No thanks” – 8 products to keep you cool at work for less than £100

Stay cool in the coming heatwave for £200 less than a Dyson fan

Architects of innovation: How ICs power the modern tech organization

Why individual contributors drive modern tech innovation

Stop chasing shadow IT. Start governing around It

Instead of chasing shadow IT, organizations should shift to real-time, policy-driven governance that operates at the moment decisions are made.
…

Could your CEO be the weakest link when it comes to AI security? New study warns execs are ‘knowingly bypassing safeguards because the perceived benefits outweigh the risks’

New research finds 62% of senior leaders use unapproved shadow AI (but only 31% of other workers), prioritizing productivity over compliance.
…

AI code security risk: The need for a smarter layer between detection and remediation

AI is accelerating code and risk. Security needs smarter triage beyond detection to remediation.