AI

Perplexity Brain redefines AI memory by focusing on agent performance and work-based context graphs rather than user data.

Tesco is migrating 40,000 server workloads away from VMware after citing 'abusive conduct' regarding pricing and support changes.

A reported multi-billion dollar cloud infrastructure deal between Microsoft and Oracle collapsed due to US government security framework requirements, highlighting compliance friction as a major factor in AI build-outs.

A critical vulnerability in Microsoft Copilot, named SearchLeak, allowed hackers to extract sensitive data, including 2FA codes, by exploiting the AI's inability to distinguish user commands from malicious instructions.

Accenture AI shopping agents earn consumer trust: 74% of consumers trust them more than friends for purchases, per 2026 Consumer Pulse Research.

Google Research's Gemini-SQL2, powered by Gemini 3.1 Pro, has achieved 80.04% execution accuracy on the BIRD Text-to-SQL Leaderboard.

A coalition of 42 state attorneys general opens a broad investigation into OpenAI, raising significant OpenAI regulatory risk days after its confidential IPO filing at an $852 billion valuation.

Google's new DiffusionGemma model promises 4x faster text generation. Here is what this experimental, parallel-processing AI means for developers and power users.

Dozens of Microsoft packages were hit with credential-stealing malware triggered by AI coding agents. Here is what to do.

The UK government partners with ElevenLabs to advance public service accessibility, AI safety research, and talent growth.

Sound Blaster Katana V2X speakers let Bluetooth attackers overwrite firmware and type commands on a PC—no pairing needed.

Meta Business Agent automates conversational commerce directly inside Instagram, Messenger and WhatsApp, collapsing the checkout funnel and forcing brands to weigh native distribution against architectural independence.