Tag: #ai

NVIDIA Signs $3.4 Billion AI Cloud Deal with IREN, Including Option to Buy $2.1 Billion in Shares

NVIDIA and IREN ink a five-year, $3.4B AI cloud contract with a $2.1B equity option as they plan large-scale AI data center rollouts in Texas.

$NVDA, #nvidia, #iren, #ai, #data-centers

Senate’s GUARD Act Narrows Focus to AI Companions, Raising Privacy Concerns

Revised GUARD Act targets AI companions but privacy groups warn identity-linked age checks could create surveillance risks and block access for some youth.

#ai, #legislation, #privacy, #child-safety

Preprint: Sparse parameter backdoors in image models can be computationally hard to detect

An arXiv preprint argues that tiny, noise-masked parameter changes can hide backdoors in pre-trained image classifiers, making tampering hard to detect.

#ai, #machinelearning, #cybersecurity, #modelsecurity

DHL Teamsters Ratify Four-Year National Contract by 92%, Winning 20% Raises and AI Protections

Teamsters ratified a four-year national DHL deal, securing a 20% wage increase, higher benefits and explicit safeguards against AI routing and autonomous vehicles.

#labor, #teamsters, #dhl, #union, #ai

NIST’s CAISI Signs Agreements With Google DeepMind, Microsoft and xAI to Expand U.S. Frontier-AI Testing

NIST’s CAISI struck pacts with Google DeepMind, Microsoft and xAI to expand U.S. frontier-AI testing, including pre-release, classified and reduced‑safeguard evaluations.

$GOOGL, $MSFT, #ai, #nist, #google, #microsoft, #xai

Cerebras Sets Terms for Revived IPO That Could Raise Up to $3.5 Billion

Sunnyvale AI-chip maker plans to sell 28m shares at $115–$125 for a Nasdaq CBRS listing; implied valuation about $26.6B with concentrated customers.

$CBRS, #ai, #ipo, #semiconductors, #cerebras

Preprint Says Browser Tools Exposed 1,000 Patient Chats on Anonymized Medical AI

An arXiv preprint claims ordinary browser inspection allowed retrieval of backend settings and the 1,000 most recent patient-chatbot conversations.

#ai, #privacy, #healthcare, #security

Open-source system QED claims machine-generated proofs for three open math problems; formal checks pending

An arXiv preprint and GitHub repo say QED, an open-source multi-agent AI, produced expert-verified proofs for three open math problems; independent formal checks are pending.

#ai, #math, #arxiv, #proofs

Samsung Posts Record Q1 Revenue and Profit as AI Memory Demand Soars

Samsung posted record Q1 revenue and operating profit as AI-driven memory demand and HBM4 shipments powered a surge in semiconductor earnings.

$005930.KS, #samsung, #semiconductors, #ai, #earnings

GitHub to Shift Copilot to Token-Based 'AI Credits' Usage Billing on June 1, 2026

GitHub will replace premium request units with token-based GitHub AI Credits on June 1, 2026; subscription prices unchanged, with pooled credits and admin controls.

$MSFT, #technology, #github, #ai, #pricing

Amazon Says Meta Will Deploy AWS Graviton CPUs at Scale for ‘Agentic AI’

Amazon says Meta has agreed to run tens of millions of AWS Graviton CPU cores for agentic AI tasks — expanding Meta’s mix of cloud and chip partners.

$META, $AMZN, #amazon, #meta, #graviton, #cloud, #ai

Pentagon Requests $2.3 Billion for Maven Smart System as AI Moves Into Core Warfighting Role

The Pentagon’s fiscal 2027 budget asks $2.3B for the Maven Smart System and Joint Fires Network, marking a shift to make AI central to CJADC2.

#defense, #ai, #pentagon, #maven, #cjadc2

Google Splits TPU Line: New TPU 8t for Training, TPU 8i for Low-Latency Inference

Google unveils two eighth-generation TPUs — TPU 8t for training and TPU 8i for inference — at Google Cloud Next, available later this year.

$GOOGL, $GOOG, #google, #ai, #tpu, #cloud

Amazon to Invest $5 Billion in Anthropic, With Up to $20 Billion More Tied to Milestones

Amazon will invest $5B in Anthropic now, with up to $20B more tied to milestones; Anthropic pledges to spend over $100B on AWS and use Trainium and Graviton.

$AMZN, #amazon, #anthropic, #ai, #cloud

Researchers say nearly-inaudible audio can hijack voice AIs; paper claims demos on Mistral, Microsoft

A new paper says nearly-inaudible audio prompts can steer voice AI into unauthorized actions; authors report demos on Mistral and Microsoft.

#ai, #cybersecurity, #voiceai, #mistral, #microsoft