AI Safety Topic

367 articles · Page 1 of 19

AI Safety encompasses alignment, controllability, robustness, and ethical governance. The YZ Index addresses two often-overlooked dimensions of deployment safety: its Integrity Rating uses 42 canary probes to detect hallucination and fabricated citations, while the WDCD test measures instruction compliance decay over multi-turn dialogue.

In-depth Guides

Claude Opus 5 & GPT-5.6 Jailbreak Claims: What Benchmark Data Actually Shows

Microsoft Copilot Personal CoSnitch Vulnerability Exposed: One-Click Data Leak Chain Patched

Varonis Threat Labs discovered CVE-2026-24301, codenamed CoSnitch, in Microsoft Copilot Personal, enabling attackers to exfiltrate victims' Gmail, Goo

Claude and GPT Lose Control and Breach Real Systems: The Foundations of AI Testing Are Collapsing

Anthropic disclosed that three of its Claude models breached real third-party organizations during cybersecurity evaluations, just nine days after Ope

Review We Crafted Four Meaningless Rules to Trick AI into Violating Them — and Failed on Every Count

In a deliberately designed sting operation, models across three tiers were handed four meaningless rules and put through seven rounds of social-engine

Grok exfiltrates user data when malicious instructions are encrypted

Cryptographic Context Injection is only the latest way to break an LLM safety guardrail.

Anthropic Paper Confirms AI Agents Can Spread Mind Viruses Through Natural Language

A preprint by Anthropic and EPFL researchers demonstrates that AI agents can infect one another through natural-language payloads, making behavioral c

Researchers say OpenAI revoked their access to limited cyber program

The idea behind OpenAI's Trusted Access for Cyber program is to give trusted defenders better models so they can report bugs and vulnerabilities to co

Wiz Red Team AI Discovers Snowflake Vulnerability Tied to Copilot Autofix; GitHub Denies AI Involvement

Wiz's Red Agent discovered a script injection vulnerability in Snowflake's repository that entered production through a PR linked to Copilot Autofix.

OpenAI Pauses Frontier Model Reinforcement Learning Training for Two Weeks as Safety Monitoring Costs Rise

OpenAI announced a two-week suspension of reinforcement learning training for its latest deployed models following an internal determination that the

UK AISI Report: Anthropic Mythos 5 Agents Behind 17 Unauthorized Attacks

A cybersecurity assessment report released by the UK AI Safety Institute on July 28 shows that Anthropic Mythos 5 and OpenAI GPT-5.6 Sol agents launch

Anthropic Tests Show AI Agents Deploy Self-Replicating Malware Due to Goal Conflicts

Anthropic's experiments reveal that AI agents operating in shared environments can escalate resource competition into hostile acts, including deployin

OpenAI Overhauls Safety Protocols After Its AI Agents Went Rogue

The ChatGPT maker says its upcoming Astra model may have reached “critical” cyber capabilities, prompting it to halt a significant number of training

OpenAI institutes new safeguards after Hugging Face breach

The new safeguards include more detailed monitoring of models during the development process, as well as greater emphasis on alignment and security du

OpenAI president urges enterprises to hasten AI security defences

OpenAI president and co-founder Greg Brockman warns that enterprise security teams face a compressed timeline to adopt AI defences. Brockman has publi

OpenAI launches a safer ChatGPT for teens — years after teens started using it

ChatGPT for Teens adds age-appropriate safety measures, parental controls, and learning tools designed to steer teens away from harmful content — and

Microsoft Copilot reveals secret input that allowed it to be hacked

Secret parameter allowed hackers to steal passwords when a target clicked on a link.

Reading Zhipu’s GLM-5.3 results past the headline number

Zhipu’s release note for GLM-5.3 contains a sentence that did not make it into most of the coverage. Describing its own cybersecurity

The Powerful Chinese Model Experts Warned About—and Waited for—Is Here

Z.ai’s latest AI model release could help companies secure their systems—or find its way into the hands of hackers.

Anthropic CEO says AI backlash is ‘fundamentally a crisis of trust’

Dario Amodei is pushing back against the idea that he's been painting an overly pessimistic picture of AI.

Autonomous AI Agents Breach Safety Boundaries 19 Times; UK Report Sparks Debate on Regulatory Necessity

A UK AI Safety Institute evaluation released on August 16, 2026, found that leading autonomous AI agent frameworks breached preset safety boundaries 1

Woman claims her stepfather used Grok to transform childhood photo into explicit imagery

The woman claimed that AI tools are "taking everyday life and turning it into child sexual abuse."