AI News
Cloudflare Unveils Advanced Security Features to Protect Data from AI Applications
The newly announced enhancements include AI Security Posture Management (AI-SPM), AI Prompt Protection, and expanded MCP Server Control.
Pritam Bordoloi
Pritam Bordoloi is a senior technology journalist and founder of The Left Shift, covering AI, enterprise technology, startups, digital transformation, and the business of technology.
Chips
D-Matrix Unveils 3D DRAM Breakthrough to Shatter AI Inference Memory Bottlenecks
The new 3DIMC architecture stacks DRAM in three dimensions, reducing latency, dramatically increasing bandwidth, and eliminating the so-called “memory wall.”
AI News
Elon Musk’s X and xAI Sue Apple and OpenAI Over Alleged AI Monopoly Collusion
The lawsuit claims Apple is using its dominant position in the smartphone market to favor OpenAI’s ChatGPT.
Leadership Moves
AHEAD Appoints Sumed Marwaha as India Managing Director
Prior to AHEAD, Marwaha served as Managing Director and Regional Vice President at Unisys India.
Cybersecurity
CloudSEK Uncovers AI Summarizer Exploit Turning Trusted Tools into Ransomware Delivery Channels
By embedding malicious payloads in HTML using CSS-based obfuscation attackers can trick AI summarizers into reproducing harmful, step-by-step ransomware instructions.
Developer's Corner
Check Point Launches Harmony SASE MCP Server, Enabling Secure AI Integration with Enterprise Networks
The Harmony SASE MCP Server is available via the Infinity Portal, with full documentation and starter queries to begin AI-assisted exploration.
Developer's Corner
Okta Launches Open-Source Auth0 Detection Catalog for Proactive Threat Monitoring
Now available on GitHub, the catalog enhances Auth0’s native threat detection capabilities.
Open Source
GitHub Open-Sources MCP Server to Improve LLM Accuracy and Automation
It acts as a source-of-truth interface between GitHub and any LLM.
Expansion
Snowflake Opens New Office in Bengaluru
The company has recently opened a new office in Mumbai.
Partnerships
Meta Partners with Midjourney to Boost AI Image and Video Tools
The announcement was made by Meta’s Chief AI Officer Alexandr Wang on Threads.
AI Startups
Anthropic Builds AI Tool to Flag Nuclear Risk Conversations
Built using a curated list of nuclear risk indicators from the NNSA, the classifier was trained and validated with over 300 synthetic prompts.
Acquisition
Databricks to Acquire Tecton, Bolstering Real-Time AI Capabilities
The acquisition will strengthen Databricks’ ability to deliver real-time, low-latency AI applications for enterprise customers.