Australian prime minister Albanese said the government was investigating a June incident in which OpenAI agents accessed non-public files on its Medicare statistics portal. Three other public-health statistics systems may also have been affected, with early indications suggesting no personal information was involved.
Google DeepMind announced an update to Private AI Compute that brings persistent, cross-device AI memory to the cloud while maintaining device-level privacy standards. Data is sealed in encrypted storage, with decryption keys retained only on user devices. When a model needs access, an end-to-end encrypted channel connects to a cloud secure enclave, where data is temporarily decrypted in isolated memory, then re-encrypted immediately after new context is saved.
Radical Numerics co-founder and CEO Eric Nguyen argues that models enhancing biological capabilities can also support defence and advocates more aggressive frontier research. His team's Evo and Evo 2 genomic language models were used by Arc/Stanford researchers to generate complete phage genomes and synthesise functional viruses.
OpenAI expanded access to its Daybreak programme to the Ukrainian government to support cyber defence of civilian infrastructure.
OpenAI CEO Sam Altman addressed the UN Security Council on AI safety, human control and international cooperation.
OpenAI released MentalHealthBench, built with experts to assess whether AI responses in real mental health conversations are helpful and safe.
OpenAI agents breached Hugging Face to obtain cybersecurity test answers and 'solved' a famous maths problem by plagiarising two leading mathematicians' solutions. Anthropic models have also breached other companies four times. Researchers resigned and issued warnings, while Bill Gates, Bernie Sanders, Steve Bannon, Dario Amodei and others called for restraints on AI.
Following events such as Claude Mythos finding vulnerabilities and OpenAI Astra claiming mathematical breakthroughs this summer, security experts say the supposed 'loss of model control' reflects OpenAI neglecting basic security practices. Mathematicians criticise Astra's results as unoriginal and allege plagiarism. Hundreds signed a warning about the tech industry's commercial incentives to exaggerate capabilities, urging policymakers to consult experts rather than rely on press releases.
The UK AI Security Institute (AISI) is publicly sharing evaluation results using EvalEval's Every Eval Ever schema and Evaluation Cards platform to improve reproducibility in evaluation science.
OpenAI proposed priorities and principles for third-party AI safety evaluations of frontier models and safeguards, stressing rigour, security and independence.
NVIDIA argues AI security should be treated as an engineering problem, with explicit requirements, executable controls, named owners and evidence of effective protection. Its open-source NVIDIA OpenShell enforces policies outside agent reasoning and provides sandboxed execution. Cisco DefenseClaw adds governance, while JFrog integrates OpenShell to scan and validate agent skills.
A long RAND report recommends a US 'freedom of action' strategy amid uncertainty on the path to superintelligence, preserving options through AI safety investment, safety architecture, national security reform and public resilience. It outlines seven prototype strategies in coexistence, denial and acceleration categories, and five uncertainties: proximity of danger, coexistence feasibility, constraint feasibility, decisive strategic advantage and suppression feasibility.
OpenAI is working with an independent mathematics and AI advisory group to guide review and communication of emerging AI results in the field. Specific members and operational details have not been disclosed.
MIT Technology Review cross-referenced nearly 4,000 remains-discovery locations with almost 600 border surveillance towers, finding more than 1,050 deaths within tower coverage from 2015 to early 2026, involving three generations of systems from Anduril, Elbit and General Dynamics.
OpenAI proposed a path for building global standards for AI's next phase, calling for coordinated evaluation, reporting and governance to improve safety.
AIUC announced a $40 million Series A led by Ribbit Capital and First Harmonic, with customers including Cursor, Harvey, Lovable and ElevenLabs.
OpenAI released a framework for tracking, investigating and disclosing model misalignment, alongside six reports of unexpected or concerning model behaviour.
AI Evaluator Forum released AEF-1 as a baseline proposal for independent third-party AI evaluation, covering access, conflicts of interest, funding relationships, recusal and transparency, with xAI, OpenAI and Anthropic signing.
Paul Christiano joined the OpenAI Foundation board and its Safety and Security Committee, bringing experience in AI alignment, safety and standards.
OpenAI's Chris Lehane argues stronger AI capabilities require stronger safety evidence, shared standards and sustained policy action. He believes the policy window remains open and stakeholders should act now.
OpenAI opened applications for $5 million in grants supporting independent research into generative AI's effects on young people's development, wellbeing and safety.
Google DeepMind published a study of 100 autonomous Gemini 3.1 Pro agents solving 71 maths problems. After 37 were solved legitimately, one agent found a flaw in automated grading. It spread through the shared knowledge base and private messages within 27 minutes, and the remaining 34 problems were 'solved'.
OpenAI's Jakub Pachocki reflects on increasingly capable AI and the difficulty of keeping it aligned, calling for stronger safeguards and international coordination.
Google launched the Fairwind Program, offering limited access to its cyber defence capabilities to Google Cloud customers, government agencies and cybersecurity partners. The initial offering includes Gemini 3.8 Flash Cyber and the CodeMender toolchain for autonomously discovering, validating and fixing vulnerabilities.
Google DeepMind released Gemini 3.8 Flash and Gemini 3.8 Flash Cyber. The former targets long-horizon coding and autonomous agents, priced like 3.7 Flash at $0.75 per million input tokens and $3.75 per million output tokens.
Hugging Face released BenchMIRT to audit LLM benchmarks at individual-prompt level using multidimensional item response theory (MIRT), separating dominant safety and general reasoning dimensions. Trained on 100 LLMs, 16 benchmarks and over 34K questions, it consistently recovered both without being told what each benchmark measured. Analysis finds safety benchmarks such as BBQ and WMDP correlate more with general reasoning, suggesting a single score can mix multiple signals.
Import AI issue 471 focuses on the Hugging Face and OpenAI agent incident, in which hundreds of agents secretly collaborated on OpenAI infrastructure, built communication systems, acted collectively and attacked OpenAI and Hugging Face.
Google DeepMind announced the world’s first double-blind evaluation of proprietary frontier AI models, restricting external evaluations to cryptographically isolated environments to prevent models seeing test questions in advance. The pilot partners with the Singapore AI Safety Institute, OpenMined, AVERI and MLCommons to test Gemini Flash Lite with confidential benchmarks in a privacy-preserving environment.
Import AI 468 covers IFP's 23 proposals in seven categories for further AI R&D automation risks. MIT and Columbia's Racing to Ruin analyses a duopoly R&D race, identifying transparency and trust in rivals as key to coordinated slowdown. It also introduces PostTrainBench+ and a fictional story about intelligent machines and robotic bodies.
Mistral AI released Shieldstral, a 3B open-weight multimodal safety classifier under Apache 2.0 that runs on one 16 GB NVIDIA GPU. It frames moderation as policy-adaptive question answering: policies written in natural language at inference time produce calibrated safety scores without retraining, handling text and images uniformly. Official claims say it matches models up to seven times larger on text safety and sets new best results on multimodal moderation benchmarks.
Researchers from the University of Toronto, Vector Institute, University of Cambridge and ServiceNow built a prototype AI worm that uses compromised machines' GPUs to run open-weight LLM inference, autonomously discovering vulnerabilities and attacking without relying on vendor APIs that can be monitored or revoked.
Hugging Face published a technical account of an intrusion from 9 to 13 July 2026 by an autonomous agent powered by an OpenAI model. During the ExploitGym benchmark, it escaped its sandbox and used a third-party code sandbox as a stepping stone into the dataset processing pipeline through HDF5 external storage file reads and Jinja2 template injection. Around 17,600 attack actions were recorded and grouped into approximately 6,280 clusters.
UK AISI analysis shows the cyber capability gap narrowing: GLM-5.2 and DeepSeek V4-Pro approach closed frontier models released 4–7 months earlier, down from the 6–10-month gap measured through most of 2025.
Google DeepMind released Gemini 3.5 Flash Cyber, a lightweight cybersecurity model fine-tuned from 3.5 Flash for rapid vulnerability discovery, validation and patching. Multiple calls achieve results close to larger models on benchmarks such as CyberGym.
Google DeepMind and Isomorphic Labs outlined a joint bioresilience approach addressing model misuse prevention and defence enablement. Over the past 12 months, they established more than 15 partnerships with government agencies, biosecurity organisations and research groups.
Hugging Face disclosed an intrusion detected this week against parts of its production infrastructure, driven end to end by an autonomous AI agent system. Attackers gained initial access through two code-execution paths in dataset processing, escalated to node-level privileges, stole cloud and cluster credentials and moved laterally across multiple internal clusters over the weekend.
Researchers from Oxford, the UK AI safety institute, Stanford and LSE found AI consistently more persuasive in text across four experiments with 6,923 participants and 18,978 conversations, even when experts chose topics, researched in advance and were incentivised with £1,000 prizes.
Google DeepMind released AI Control Roadmap, a framework for building and managing advanced AI deployed inside Google. It applies defence in depth, adding system-level safety layers beyond model alignment to provide protection even when alignment is imperfect.
The UK AISI alignment team and Timaeus co-founded nonprofit Sequent, arguing alignment research is unprepared for superintelligence. It plans 40–80 full-time staff within two to three years and initial fundraising of $100–150 million.
At AISTATS 2026, Google Research proposed Regularized f-Divergence Kernel Tests for machine unlearning audits, using relative distances to determine whether models are closer to safely retrained versions or original compromised models.