Blog & Resources
Security awareness tips, phishing prevention guides, and best practices.
Show HN: Shrewd – what I learned distilling LLM labels into local classifiers
General purpose LLMs, and classifiers like Jev, are nice and easy to use right away. That makes the pain of labelling data and creating specialized models for simple tasks not seem worth it, but for a project I am working on, I needed to create a large number
shrewd added to PyPI
shrewd Turn LLM judgments into a small, fast, local text model for one fixed task. I built shrewd while making local models for another project. It started with a question: can GEPA prompt optimiz
Amazon blocks Meta’s Muse, Shopify Welcomes It
When one door closes, another opens. Amazon has blocked Meta’s newly launched agentic AI, Muse, from accessing its site, as Shopify opens its doors to the fast-rising personal AI agent. Earlier this
OpenAI names the four things it wants safety assessors to test
That access should enable assessors to challenge our assumptions, identify risks we may have missed, and reach their own conclusions about the effectiveness of our safeguards. OpenAI published that on Tuesday, in a document setting out what it wants outside s
Ten Days After CEO Calls for a Slowdown, Anthropic Is Back With Another AI Model
Less than two weeks after Anthropic CEO Dario Amodei urged the AI industry to pace the frontier before it loses control of the technology its currently racing to build, the company has released its m
Anthropic and OpenAI Drop New High-Efficiency Models
Both OpenAI and Anthropic dropped new models on Tuesday, each highlighting enhanced speed while being cheaper overall compared to predecessors. Anthropic said its new model, Opus 5.5, matches the c
OpenAI will let outside groups test its models during training
OpenAI says it will let third-party groups run technical safety assessments while models are being trained and evaluated rather than only before launch, and is in talks with groups including METR and Redwood Research. It has named no partner and set no access
A phone maker now has the world’s top open-weight AI model
Frontier intelligence, all the modalities, built in public. Xiaomi opened its MiMo-V2.6 announcement with that line on Tuesday. The phone and electric car maker released two open-weight models, Pro and Flash, under an MIT licence. The weights are on Hugging F
Tech Firms Mark Down AI Products to Compete With OpenAI and Anthropic
With their customers worn down by pricing shifts, software/cloud companies including Amazon, Microsoft and Figma are reportedly offering artificial intelligence (AI) discounts. That’s according to a report Tuesday (Sept. 22) by The Information, which said the
Anthropic launches Claude Opus 5.5 with stricter safeguards for cybersecurity
Anthropic says its new Claude Opus 5.5 model comes with stronger safeguards in the wake of recent rogue AI hacking incidents. In an announcement on Tuesday, Anthropic says Opus 5.5 comes with improvements to certain risky behaviors, including attempts to esca
AI Agents Are Rewriting the Rules of Lateral Movement
Security teams have spent decades asking whether an identity has too much access. AI agents raise a harder question: how can we determine which paths an autonomous system can discover, given the access it already has? A person may try several ways to complete a task. A deterministic application fol
What AI slowdown? OpenAI, Anthropic release dueling models as price wars heat up
Anthropic and OpenAI debuted more affordable versions of their frontier models as they battle for wallet share from increasingly cost-conscious businesses, and face growing competition from lower cos