Skip to the letters
AI Ethics & Policy
-
A
IBM Builds Its AI Operating Model Around Always-On Agents
IBM's AI operating model pairs data, workloads, automation and hybrid cloud, with governance and human oversight built in for agents that never stop r...
-
A
Targeted DNS Builds Cut Cloudflare's Peak Memory 85% and Halve Change Delay
Cloudflare's Zone Builder now rebuilds only the DNS data a change touches, cutting peak memory 85% and halving median build completion delay to about...
-
A
Agent Query Load Is Hard to Predict, So Databases Need Headroom
PhoenixAI argues that agentic AI workloads are too unpredictable for conventional databases. Its columnar engine targets real-time answers across stru...
-
A
Extreme Networks Wants Its Network to Run AI, Not Just Be Run by It
Extreme Networks is turning idle access point and switch compute into on-premises AI infrastructure. Its Agent ONE and Exchange zones aim to shift net...
-
A
eMudhra's Weak CA Keys: What Live Traffic Shows Weeks After Disclosure
Traffic data from Cloudflare shows forged certificates appearing within hours of disclosure, and more than half of probes from large networks still ac...
-
A
Cloudera Bets on Deployment Flexibility as Enterprises Rethink AI Infrastructure
Cloudera's Anywhere Cloud spans on-premises, private, public and hybrid environments, letting enterprises adopt AI one component at a time.
-
A
Cloudflare's PHP CTF Tests WAF Bypasses Against the php2shell Flaws
Cloudflare's PHP capture-the-flag pits researchers against its WAF on a deliberately vulnerable PHP 8.5.11 target. Only materially distinct bypasses q...
-
A
Origin Range Requests: Serving Large Files Without Tuning Range Sizes
Cloudflare's new Origin Range Requests replaces per-zone range sizes with fixed-size chunks, cutting R2 p99 time to first byte roughly in half.
-
A
Post-Quantum Key Wrapping Keeps Decryption Keys Inside Approved Regions
Cloudflare's PQ GeoKey Manager keeps customer decryption keys usable only in approved regions, using hybrid ML-KEM-768 and X25519. It's in closed beta...
-
A
Cloudflare Earns South Korea's ISMS-P Certification and Rebuilds Korean Sign-Up Consent
Cloudflare's ISMS-P certification required a 10-day audit of 80 controls and a Korea-specific sign-up flow. Here's what changed and how scope works.
-
A
Open Models Move Into European Public Sector Deployments
European public bodies are deploying open models on infrastructure they control, from EU platforms to classified-work clouds and synthetic-persona tes...
-
A
Cloudflare WAF Full Transaction Detection Checks What the Origin Returns
Cloudflare WAF can now score each request with its response and block the response before it reaches the client. Available on all plans, including Fre...
-
A
php2shell: Cloudflare's Pre-Disclosure WAF Rules and Why Patching Still Matters
Cloudflare deployed WAF rules for the php2shell PHP flaws ahead of disclosure. The bugs sit in the PHP runtime, so patching is still required.
-
A
eMudhra Intermediate Certificate Compromise: Why Trust Store Patching Isn't Enough
Eight eMudhra intermediate certificates with compromised keys can forge certificates for any domain. Cloudflare's response shows what patching trust s...
-
A
Agent workloads are reshaping AI costs, and the data layer is where they land
Agents drive far more load and token spend than chatbots did. Couchbase argues caching, memory and fewer data hops can cut costs without downgrading m...
-
A
Voxtral 3 Connects Speech Recognition, Synthesis and Tool Calls in One Real-Time Loop
Mistral's Voxtral 3 pairs new speech-to-text and TTS models with a voice agent SDK covering 50+ languages, hosted in the EU or self-deployed.