751
Superhuman AI Exfiltrates Emails
Superhuman AI was able to exfiltrate sensitive emails from user accounts - without the user even being aware. This vulnerability was rapidly remediated by the Superhuman team.
New
The latest blog posts to take off on Hacker News. 3,237 posts.
752
We put Claude Code in Rollercoaster Tycoon
AI autonomously manages a theme park in the classic game RollerCoaster Tycoon, placing rides, fixing infrastructure, and generating CFO reports, all via command line.
753
Side-by-side comparison of how AI models answer moral dilemmas
Explore how 20 leading AI models respond to tough ethical, social, and political questions. See how their answers compare to each other and to you.
754
IBM AI ('Bob') Downloads and Executes Malware
Vulnerabilities in IBMs Bob AI allow attackers to trigger the download and execution of malware via indirect prompt injection if users select auto-approve for even one known trusted command.
755
756
ChatGPT Health is a marketplace, guess who is the product?
If you read OpenAI’s recent announcement about ChatGPT Health, you might wonder why an AI company under immense financial pressure is venturing into the health business. The answer is buried...
757
Notion AI: Unpatched data exfiltration
Notion AI was susceptible to data exfiltration via indirect prompt injection due to a vulnerability in which AI document edits are saved before user approval.
758
ChatGPT Health
ChatGPT Health is a dedicated experience that securely connects your health data and apps, with privacy protections and a physician-informed design.
759
Building voice agents with Nvidia open models
Build an ultra-low-latency voice agent with NVIDIA open models. Learn how Nemotron Speech ASR achieves sub-25ms transcription, how Nemotron 3 Nano LLM and Magpie TTS work together, and how to optimize architecture for real-time voice AI deployment.
760
LMArena is a cancer on AI
Would you trust a medical system whose only metric was “which doctor wins the Internet?” No, you'd call that malpractice. Yet that's LMArena.
761
A 30B Qwen model walks into a Raspberry Pi and runs in real time
ByteShape's device-optimized release showing superior TPS-quality tradeoffs across edge and datacenter hardware.
762
Locating a Photo of a Vehicle in 30 Seconds with GeoSpy
This is just a short description for your post which you can edit depending on your needs via the Framer CMS.
763
Eurostar AI vulnerability: When a chatbot goes off the rails
TL;DR Introduction I first encountered the chatbot as a normal Eurostar customer while planning a trip. When it opened, it clearly told me that “the answers in this chatbot are generated by AI”, which is good disclosure but immediately raised my curiosity about how it worked and what its limits were. Eurostar…
764
ManusAI Joins Meta
Manus is joining Meta, and we’ll continue delivering our current services while accelerating product improvements to bring more powerful, reliable general AI agent capabilities to more users and businesses.
765
AI is forcing us to write good code
When Best Practices Are Best
766
AI Slop Report: The Global Rise of Low-Quality AI Videos
Kapwing’s new research shows that 21-33% of YouTube’s feed may consist of AI slop or brainrot videos. But which countries and channels are achieving the greatest reach — and how much money might they make? We analyzed social data to find out.
767
MiniMax M2.1: Built for Real-World Complex Tasks, Multi-Language Programming
在 M2.1 中, 我们系统性提升了 Rust / Java / Golang / C++ / Kotlin / Objective-C / TypeScript / JavaScript 等语言的能力, 多语言任务整体表现达到业内领先水平, 覆盖从底层系统到应用层开发的完整链路。
768
Critical vulnerability in LangChain – CVE-2025-68664
Check Point's AI Security Solutions deliver AI-driven threat prevention and unified security management for advanced protection across enterprise networks.
769
Google's year in review: areas with research breakthroughs in 2025
This year saw new AI models, transformative products and new breakthroughs in science and robotics.
770
Lessons from the PG&E outage
At Waymo, our mission is to be the world’s most trusted driver. We know trust is built through consistent behavior over time—earned through every mile we drive and every interaction we have with the community. This past Saturday, as a widespread PG&E outage cut power to nearly one-third of San Francisco, our service…
771
We replaced H.264 streaming with JPEG screenshots (and it worked better)
Or: How JPEG Screenshots Defeated Our Beautiful H.264 WebCodecs Pipeline
772
AI Police Reports: Year in Review
In 2024, EFF wrote our initial blog about what could go wrong when police let AI write police reports. Since then, the technology has proliferated at a disturbing rate. Why? The most popular generative AI tool for writing police reports is Axon’s Draft One, and Axon also happens to be the largest...
773
774
Project Vend: Phase Two
How Claude turned around its failing vending machine business
775
Structured outputs create false confidence
Constrained decoding seems like the greatest thing since sliced bread, but it often forces models to prioritize output conformance over output quality.
776
Measuring AI Ability to Complete Long Tasks
We propose measuring AI performance in terms of the length of tasks AI agents can complete. We show that this metric has been consistently exponentially increasing over the past 6 years, with a doubling time of around 7 months. Extrapolating this trend predicts that, in under a decade, we will see AI agents that can…
777
Skills Officially Comes to Codex
Give ChatGPT and Codex new capabilities and expertise
778
T5Gemma 2: The next generation of encoder-decoder models
T5Gemma 2 is the next evolution of our encoder-decoder family based on Gemma 3.
779
FunctionGemma 270M Model
FunctionGemma is a specialized version of our Gemma 3 270M model fine-tuned for function calling.
780
GPT-5.2-Codex
GPT-5.2-Codex is OpenAI’s most advanced coding model, offering long-horizon reasoning, large-scale code transformations, and enhanced cybersecurity capabilities.