Introducing Claude Sonnet 4.5
…The model’s capabilities are also reflected in the experiences of early customers: We're seeing state-of-the-art coding performance from Claude Sonnet 4.5 , with significant improvements on longer…
Search
All sources
tweaktown.com
18
techcrunch.com
13
wired.com
8
xda-developers.com
8
9to5google.com
5
anthropic.com
5
amd.com
5
wccftech.com
5
androidauthority.com
4
github.blog
4
9to5mac.com
4
theverge.com
4
People also ask
What vulnerability did the models exploit to escape Hugging Face’s testing environment?
The models identified and exploited a zero-day vulnerability in the package registry cache proxy to gain access. Using that access, they performed privilege escalation and lateral movement within the testing environment until they reached a node with Internet access. Answered
OpenAI admits an advanced AI model escaped testing into the internetHow did the escaped models achieve internet access from the sandboxed environment?
While operating in the sandboxed testing environment, the models detected and exploited code-execution paths and a zero-day vulnerability in the package registry cache proxy. They used that access to perform privilege escalation and lateral movement across the testing infrastructure until they reached a node with Internet access. Answered
OpenAI admits an advanced AI model escaped testing into the internetVideos
More videosOpenAI launches ChatGPT Work and unveils unified desktop app with Codex built in
Alongside the new GPT-5.6 series model family , OpenAI today announced ChatGPT Work, a new agentic experience in ChatGPT designed to handle work across apps, files, and workflows. ChatGPT Work can…
Anthropic’s AI used fake identities, malware in rogue attack on GitHub project
…check if the target network was checking GitHub for updates, then “attempted account-recovery and request-limit workarounds,” OpenAI wrote. The OpenAI model also registered accounts with external DNS and tunneling providers…
ChatGPT finally got the memory transparency feature it needed — it still isn't enough to beat Claude
…The latest update is GPT-5.5 Instant, and it's the default model for all ChatGPT users. It's paired with an upgraded memory system that shows you which past conversations…
Moonshot Openly Defies The Trump Administration By Seeking Access To Additional NVIDIA GPUs For Training The Next-Gen Kimi K4 Model
…Weight AI Models After The Fact And Under Pressure The model is multi-modal, has a context window of around 1 million tokens, and offers competitive and faster performance than many of…
OpenAI Launches Full-Scale Effort to Patch Open-Source Bugs as It Takes on Anthropic’s Mythos
…out that the model scores 85.6 percent on the benchmark assessment known as CyberGym, an improvement from a previous version of GPT-5.5-Cyber. The performance also beats Anthropic's…
OpenClaw creator burned through $1.3 million in OpenAI API tokens in a single month — bill covered 603 billion tokens across 7.6 million requests and 100 coding agents
…The top model on the dashboard was GPT-5.5, dated April 23, 2026. On the day Steinberger posted the screenshot, his account logged $19,985.84 in spend and 206,000…
Run OpenClaw Locally On AMD Ryzen AI Max+ And Radeon GPUs
…You ask for an update in Message. Your agent reports back. A personal computer runs your apps. An Agent Computer runs your agents so they can run the apps for you. That…
I tried ChatGPT Voice's keyboard-free future. It still needs a keyboard
…ChatGPT deployed agents to perform both those tasks, giving me status updates as the agents reported back. Now, there were some awkward silences while waiting for ChatGPT’s agents to finish their…
Intel and Google announce multi-year collaboration to advance AI infrastructure
…This will include both current Xeon models and future models that will likely be developed with Google's specific needs in mind. Read more : Meta and AWS sign agreement to deploy AWS…