ai-security
66 posts tagged ai-security.
- What Your Agent Evaluation Harness Can Actually See
- Near AI’s breach shows where agent privacy really breaks
- AI-writing detectors do not belong in Reddit moderation by default
- Claude Code's 'Don't Ask Again' and the 1,122-file password leak
- Payment Skimming Hits Restaurant Delivery: What the Campaign Tells Us About AI-Era Web Security
- Stolen NVIDIA pallets are a compute supply chain warning
- AI exploit generation is getting cheap enough to change patch work
- The OCC puts AI agent autonomy on the bank examiner checklist
- What OpenAI's new comms chief and bigger bug bounties say about its priorities
- Claude-written Linux exploit puts patch lag in the spotlight
- Claude, game reverse engineering, and the product boundary problem
- Malware using LLMs for victim triage is the practical AI security problem
- RAMBO and the Air-Gap Myth: What RAM Radio Leaks Mean for AI Security
- A $6.3B Binance hack claim needs attribution, not vibes
- ChatGPT Account Takeovers: What a Hacker News Claim Does and Doesn't Prove
- TextSeal2 keeps watermarks alive under speculative decoding
- ChatGPT browsing access is now part of site visibility
- What OpenAI's Misalignment Detection Post Actually Describes
- TopKLoRA turns a poisoned fine-tune into a circuit you can cut out
- Smart TV reverse engineering exposes the local AI safety gap
- Sweden’s AI block is a procurement lesson, not a model verdict
- What OpenAI's Training Containment Actually Protects Against
- SERP network data is becoming an SEO research surface
- When AI Agents Start Finding Bugs Faster Than Maintainers Can Patch Them
- The DeepSeek ban is really a dependency inventory problem
- The Opencode breach report is a key-rotation drill
- When the Vulnerability Scanner Becomes an Attack Planner
- Agent red teams need maps, not just scary prompts
- Zcash’s AI privacy problem is address reuse, not magic hacking
- What Sophos actually automated to cut threat response from 38 minutes to 89 seconds
- Crypto’s AI encryption scare is really a migration planning problem
- AI-assisted breaches are becoming an operations problem
- Co-Evolving the Attacker: How AdvSim2Real Hardens Web Agents Against Prompt Injection
- AI meetups make a bad trust layer for crypto
- Chainalysis’s AI tracing claim is a workflow story, not magic attribution
- LLM security starts where code enters the pipeline
- KaliBench Measures Whether LLMs Can Actually Type the Right Security Command
- Robinhood’s trading agents make permissioning the product
- Reinforcement learning weakens reasoning distillation defenses
- Australia’s AI agent hearing is really about incident disclosure
- Agent control failures are moving from demo risk to operating risk
- NSA’s reported AI testing spend makes evals look like infrastructure
- When Agents Hack: What the OpenAI-Hugging Face Story Actually Tells Builders
- Australia’s OpenAI agent breach is a disclosure problem, not a sci-fi one
- Authorship verification works better as similarity, not classification
- Known puts domain identity into the agent stack
- Gemini’s reported company breaches show the AI disclosure gap
- VVV Shows the Privacy AI Pitch Has Become a Token Story
- Marc van der Chijs’s AI banking warning needs plumbing details
- A vulnerability is not fixed because an AI bot saw it
- The RubyGems agent report is a supply-chain warning
- An 86% Bitcoin quantum benchmark cut is not an 86% attack
- OpenAI’s Python SDK gets key expiration controls, not just new image hooks
- Harmony’s reported shutdown plan makes AI a chain-risk argument
- AI cyber risk is moving from demo to preparedness
- AI bug reports are now a Lightning ops problem
- Bitcoin’s AI Security Problem Is Scale, Not Superintelligence
- Treat the Texas AI hacking story as a disclosure systems test
- Prompt Injection Is SEO Spam With a Larger Blast Radius
- Payward’s reported Glasswing access is an AI security story, not a crypto story
- VICBench shows vulnerability detection still needs humans
- AI security review is hitting Bitcoin repos, not just toy code
- Apple’s AI slop filter may be catching real macOS bugs
- Agent immunity is the missing layer between alignment and tool use
- Promptware Treats Prompt Injection Like an Execution Chain
- Agent safety belongs outside the agent process