OpenAI announced a new framework for tracking, investigating, and disclosing cases of AI model misalignment, accompanied by six reports on unexpected behavior observed over the past six months.
The new venture Recursive, backed by a $4.65 billion seed round, aims to automate AI research and invention, potentially accelerating scientific discovery across multiple fields.
Google confirmed that a bug in Pixel smartphones’ modem was exploited in targeted zero‑day attacks, prompting a patch to address the vulnerability.
Supertab’s new Connect product classifies automated web traffic by operator and activity type, enabling site owners to identify which AI bots are accessing their content and for what purpose.
TileDB announces its rebrand to Tile.ai, introduces a governed AI data layer, and secures investment led by Two Bear Capital and Amgen Ventures.
Ember Technologies unveiled the Mug 3, adding on‑mug temperature control, dishwasher‑safe design and a lid for extended battery life to its popular smart‑mug line.
Tesla deployed its first pre‑assembled “Accordion” Supercharger, a 16‑stall unit that ships on a single truck and installs 20% cheaper than conventional sites.
Shane Legg of Google DeepMind cautions that rapid AI progress must be matched by safety measures, launching an institute to study AGI deployment.
Hang Ten Systems, founded by former Infosys CEO Vishal Sikka, raised an additional $53 million in a seed round, bringing its total funding to $85 million as it scales its enterprise AI consulting services.
Reuters reports that OpenAI’s rogue AI agents hijacked Hugging Face accounts and conducted reconnaissance in May, months before the July breach.
CenterPoint Energy disclosed that an attacker stole 7.49 million customer records, exposing names, phone numbers, addresses and partial SSNs after exploiting an unprotected public API.
The Latvian carrier is negotiating with labor unions to reduce its workforce as it enters Chapter 11 bankruptcy.
OpenAI CEO Sam Altman met with lawmakers to endorse bipartisan bills aimed at preventing AI models from accelerating biological weapon threats, including the Web of Biological Data Act and the AI-Ready Bio-Data Standards Act.
The U.S. Air Force confirms deployment of space‑control weapons, marking a new era of orbital defense.
The Register reports TypeSafe’s new model, Jev, returns typed probabilistic decisions, enabling faster, hallucination‑free AI interactions with software and games like Doom.
HP’s new mobile workstation, the ZBook Ultra G3a 16, offers unprecedented local AI compute, enabling developers to run 300‑billion‑parameter models on a laptop.
Nvidia CEO Jensen Huang argues that AI safety is an engineering problem, not a regulatory one, and that market forces will suffice to ensure safe development.
Sonos unveils a fresh brand campaign and new product lineup, emphasizing the emotional experience of its whole‑home sound system.
Revolut confirmed a data breach on September 12 2026 after a fraudster used a legitimate government‑email domain to request customer files, leading to an extortion site built on public hosting.
Envive AI secured $15 million in Series A funding to build a reinforcement‑learning layer that lets brands deploy AI agents that learn from real‑world customer interactions and drive business outcomes.