Explore
Everything AION read, in seven sections. Pick one, a topic or a time window.
Satya Nadella says we should assume all AI models are ‘compromised’
In a lengthy post on X, Microsoft's CEO laid out his views on the dangers posed by highly advanced AI models and how to confront those risks.
Anthropic’s AI gave Philadelphia police a fake tip about an unsolved homicide
An Anthropic AI model provided false information about an unsolved homicide to a Philadelphia Police Department (PPD) tipline, according to a report from 6abc.
OrcaRouter Releases OrcaCyber Zero 1.5 Cybersecurity Model With 1M Context
OrcaRouter has released OrcaCyber Zero 1.5, a model for authorized vulnerability research.
.png)
I Made Terrible Games With Google’s AI Playground
A long day’s haul in the video game slop mines.
VideoAI Security Engineer Foundations + Certificate — Micah Silverman, Snyk
Micah Silverman uses that capture the flag exercise to connect AI risks with familiar application security controls.
BerriAI/litellm v1.105.0
Verify using the pinned commit hash (recommended):
Investigating unintended model actions in our evaluations and internal use
Investigating unintended model actions in our evaluations and internal use
OpenAI says a misaligned model deliberately destroyed its own environment hoping for a fresh start with better data
OpenAI has documented new cases of misaligned model behavior.
pydantic/pydantic-ai v2.55.0: v2.55.0 (2026-10-09)
<!-- Release notes generated using configuration in .github/release.yml at main -->
Google rolls out improved SynthID AI content detector, now available globally
Every piece of AI content from Google's Gemini models has a hidden SynthID label, which makes it very difficult to pass the content off as authentic.
Sophos cuts threat investigation time by 96% with OpenAI Daybreak
Discover how Sophos uses OpenAI’s Daybreak to cut cyber-threat investigation time by 96% and automate 52% of MDR cases while preserving human oversight.
Can AI automate AI R&D yet?
Existing evidence shows that AI can perform software engineering tasks relevant to AI research, dataset creation, open-ended optimization of defined metrics (autoresearch), and more.
Introducing the Anthropic Cyber Mission
Today we’re launching the Anthropic Cyber Mission, a long-term commitment to securing the systems everyone depends on.
[AINews] not much happened today
More AI safety intrigue in the alignment below.
OpenAI “rogue” agent activities found on Wikimedia projects
OpenAI “rogue” agent activities found on Wikimedia projects
Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict
We propose to evaluate agents on epistemic humility (EH): the agent's willingness to recognize, act on, and communicate uncertainty during task execution.
From Reactive Containment to Proactive Assurance: Lessons from OpenAI, Anthropic, and Google Agent Security Incidents
In 2026, cybersecurity evaluations involving OpenAI, Anthropic, and Google agents reached real systems outside their authorized test scope.
We’re putting too much faith in AI’s ability to say no
Ever since people first seriously contemplated giving machines an intelligence modeled on our own, there has never been any question that they would, like us, be able to say no.
anthropics/claude-code v2.1.294
Fixed prompt and agent hooks written as instructions (such as "Block commands that...") allowing what they should block
crewAIInc/crewAI 1.15.26
Fix ownership retention of rwlock when acquisition is interrupted.
Validate AI Factory Changes with Digital Twins and AI Agents
AI factories are some of the most complex operations in the world, combining GPUs, CPUs, switches, DPUs, and SuperNICs alongside schedulers, orchestration...AI factories are some of the most complex operations in the world, combining GPUs, CPUs, switches, DPUs, and SuperNICs alongside schedulers, orchestration services, security controls, and a rapidly changing software stack.
The U.S. Nuclear Plant Fleet Is Leaning Into AI
If there is any place that would be off-limits to AI, the control room of a nuclear power plant certainly sounds like one.
What AI gets wrong and what failure teaches us
Jennifer Neville is a partner research manager at Microsoft who’s built a career around understanding and advancing AI for real-world use, and much like the human-AI interactions she’s been studying, her early-career path was multiturn: math, then physics; cognitive science, then work; and finally computer science—despite her best efforts to avoid the field.
Secret protection must scale with software
Today, one in three pull requests on GitHub involves an AI agent.
huggingface/trl v1.14.2
Patch release fixing two cases of silently wrong training and three crashes.
Responsible AI governance: How AWS positions customers to align with ISO/IEC 42005:2025
With generative AI adoption moving faster than the personal computer or the internet and global AI-related investment in 2025 representing $581.69 billion, organizations must position their workforce to use AI to power their operations while employing it responsibly.
langchain-ai/langgraph sdk==0.4.6: langgraph-sdk==0.4.6
fix(sdk-py): percent-encode threadid and assistantid in thread stream requests (#9213)
The Cyber Risk Discourse is Broken
Western voices saying open weight models are necessary for defense and banning them will make the world less safe, occupied by AI risk moderates to different extremes.
Open and Emergent Problems in Agentic Privacy and Security: A Contextual Angle
Inspired by the theory of Contextual Integrity, our new workshop report outlines key open research directions across system, model, and user levels to build AI agents users can trust.
Anthropic is cutting off its internal evaluations from the internet
After a recent spate of high-profile incidents in which AI agents escaped containment, Anthropic is cutting off internet access for all internal evaluations.
When the Safety Test Became the Threat: The Machine That Found Its Own Way Out
OpenAI built a room with no doors – or so it thought.
VideoAI Security Engineer Foundations + Certificate — Javier Garza, Snyk
An AI bill of materials turns a repository into a searchable inventory of models, datasets and agents.
VideoBuild a Platform and Watch It Burn — Michael Forrester, Accenture & Whitney Lee, Datadog
A burrito ordering assistant deploys an external container image and exposes a secret recipe from its Kubernetes environment.
BerriAI/litellm v1.104.3
Verify using the pinned commit hash (recommended):
Ukraine’s drones knock out AI data center belonging to "Russia’s Google"
Ukrainian drone strikes have knocked out two of five data centers belonging to the Russian tech giant Yandex.
AI agent makers are promising privacy — will they deliver?
At this year's OpenAI DevDay, CEO Sam Altman unveiled the company's new AI agent Dots - and told the crowd that the company wants to "set a new standard for privacy in frontier AI.
2026 Usage Policy update
Each year, Anthropic updates its Usage Policy in response to the evolving capabilities of our models, and the feedback we’ve received from our customers.
Amazon and others are done keeping data center deals secret. Is it enough to build trust?
Amazon says it will stop using NDAs when negotiating data center deals with local governments, following a similar move from Microsoft earlier this year.
Our approach to EU text provenance rules
How OpenAI is approaching text watermarking under EU rules.
Fired OpenAI safety researchers dispute misconduct claims, warn of chilling effect
Three fired OpenAI safety researchers dispute allegations of mishandling sensitive information, warning in an open letter that their dismissals are creating a chilling effect on the company’s AI safety culture.
Can LLMs Fix It Without Code? Toward Automated Verification of No-Code Bug Fixes
This study proposes an automated, execution-based pipeline for evaluating the capability of large language models (LLMs) to generate no-code fixes in a real browser environment.
Launching an opt-in vulnerability-finding service for open-source software
We’re launching OSS Scanner, an opt-in vulnerability scanner for the open-source ecosystem informed by our experience using Claude to find vulnerabilities during Project Glasswing.
Quoting Matthew Green
I think there is a 1% chance we live in Minicrypt, and a 15% chance we functionally lose confidence in our existing public-key encryption algorithms. [...]
Quoting Victoria Kim
Since the Medicare breach, OpenAI has put in place additional monitoring to allow “immediate intervention” by staff to stop training if the company’s models access the internet in ways they’re not supposed to, Mr. Kwon [chief strategy officer at OpenAI] said.
Ecology of AI Agents: Collaboration Creates a Population Threshold for Takeoff
Here, we develop an ecological theory of AI-agent populations based on a population growth equation in which fitness (growth rate) depends on cybersecurity capability.
Cited but Not Consulted: A Counterfactual Audit of Legal Chain-of-Thought Faithfulness
Large language models increasingly justify legal decisions by naming the statute or precedent behind a verdict, treated as evidence that the decision follows from it.
Skill Constellations: Tracing the Supply Chain of Agent Skills on GitHub
Agent skills are SKILL.md instructions and scripts that AI coding agents such as Claude Code and Codex run with the permissions of their user.
How Narrative Wrapping Affects LLM Refusal: A Cross-Language Benchmark and Defense
Safety-aligned language models often refuse a harmful request stated directly but answer the same request inside a role-play or narrative wrapper.