A recent experience with ChatGPT 5.5 Pro https://gowers.wordpress.com/2026/05/08/a-recent-experience-with-chatgpt-5-5-pro/
Gowers's Weblog
A recent experience with ChatGPT 5.5 Pro
We are all having to keep revising upwards our assessments of the mathematical capabilities of large language models. I have just made a fairly large revision as a result of ChatGPT 5.5 Pro, to whi…
Using Claude Code: The unreasonable effectiveness of HTML https://twitter.com/trq212/status/2052809885763747935
X (formerly Twitter)
Thariq (@trq212) on X
Using Claude Code: The Unreasonable Effectiveness of HTML
EU Parliamentary Research Service calls VPNs "a loophole that needs closing" https://cyberinsider.com/eu-calls-vpns-a-loophole-that-needs-closing-in-age-verification-push/
CyberInsider
EU calls VPNs “a loophole that needs closing” in age verification push
The European Parliamentary Research Service has warned that VPNs are increasingly being used to bypass online age-verification systems.
Internet Archive Switzerland https://blog.archive.org/2026/05/06/internet-archive-switzerland-expanding-a-global-mission-to-preserve-knowledge/
LLMs corrupt your documents when you delegate https://arxiv.org/abs/2604.15597
arXiv.org
LLMs Corrupt Your Documents When You Delegate
Large Language Models (LLMs) are poised to disrupt knowledge work, with the emergence of delegated work as a new interaction paradigm (e.g., vibe coding). Delegation requires trust - the...
I’ve banned query strings https://chrismorgan.info/no-query-strings
chrismorgan.info
I’ve banned query strings — Chris Morgan
Idempotency is easy until the second request is different https://blog.dochia.dev/blog/idempotency/
blog.dochia.dev
Idempotency Is Easy Until the Second Request Is Different | Dochia CLI Blog
Idempotency is not just an HTTP header or a key lookup. This article covers the failure cases that bite real APIs: different requests with the same key, concurrent retries, partial success, downstream uncertainty, response replay, expiry, and duplicate message…
I returned to AWS and was reminded why I left http://fourlightyears.blogspot.com/2026/05/i-returned-to-aws-and-was-reminded-hard.html
Blogspot
I returned to AWS - and was reminded HARD why I left.
I was one of the very first advocates for AWS back when it was brand new - SQS, S3, EC2 SimpleDB - it was a lot smaller back then. In fact I...
Hardware Attestation as Monopoly Enabler https://grapheneos.social/@GrapheneOS/116550899908879585
GrapheneOS Mastodon
GrapheneOS (@GrapheneOS@grapheneos.social)
Apple and Google are gradually expanding their use of hardware-based attestation. They're convincing a growing number of services to adopt it. Google's Play Integrity API and Apple's App Attest API are very similar. Apple brought it to the web via Privacy…
Local AI needs to be the norm https://unix.foo/posts/local-ai-needs-to-be-norm/
unix.foo
Local AI Needs to be the Norm · unix.foo
Local AI models should be the default.
🤡2👏1
Training an LLM in Swift, Part 1: Taking matrix mult from Gflop/s to Tflop/s https://www.cocoawithlove.com/blog/matrix-multiplications-swift.html
Cocoa with Love
Training an LLM in Swift, Part 1: Taking matrix multiplication from Gflop/s to Tflop/s | Cocoa with Love
10 implementations of handwritten matrix multiplication: from plain C and Swift through to Metal
Nullsoft, 1997-2004 (2004) https://slate.com/technology/2004/11/the-death-of-the-last-maverick-tech-company.html
Slate Magazine
The death of the last maverick tech company.
When America Online purged its tiny Nullsoft branch of all but three employees this week, it lost arguably the most prolific division of the...
Interfaze: A new model architecture built for high accuracy at scale https://interfaze.ai/blog/interfaze-a-new-model-architecture-built-for-high-accuracy-at-scale
Interfaze
Interfaze: A new model architecture built for high accuracy at scale
A complete walkthrough of Interfaze: what it is, who we benchmark against (Gemini-3-Flash, Claude-Sonnet-4.6, GPT-5.4-Mini, Grok-4.3, plus task specialists like Reducto, SAM 3, Scribe v2), full results across 9 benchmarks, and code examples for OCR, object…
UCLA discovers first stroke rehabilitation drug to repair brain damage (2025) https://stemcell.ucla.edu/news/ucla-discovers-first-stroke-rehabilitation-drug-repair-brain-damage
UCLA BSCRC
UCLA discovers first stroke rehabilitation drug to repair brain damage
A new study by UCLA Health has discovered what researchers say is the first drug to fully reproduce the effects of physical stroke rehabilitation in model mice.
Library for fast mapping of Java records to native memory https://github.com/mamba-studio/TypedMemory
GitHub
GitHub - mamba-studio/TypedMemory: A Java 25 library for mapping records to strongly typed off-heap memory using the FFM API.
A Java 25 library for mapping records to strongly typed off-heap memory using the FFM API. - mamba-studio/TypedMemory
GitLab announces workforce reduction and end of their CREDIT values https://about.gitlab.com/blog/gitlab-act-2/
about.gitlab.com
GitLab Act 2
A letter to our customers and our investors.
Postmortem: TanStack NPM supply-chain compromise https://tanstack.com/blog/npm-supply-chain-compromise-postmortem
Tanstack
Postmortem: TanStack npm supply-chain compromise | TanStack Blog
On 2026-05-11, an attacker chained a pull_request_target Pwn Request, GitHub Actions cache poisoning across the fork↔base trust boundary, and OIDC token extraction from runner memory to publish 84 malicious versions across 42 @tanstack/* packages on npm.…
Google says criminal hackers used AI to find a major software flaw https://www.nytimes.com/2026/05/11/us/politics/google-hackers-attack-ai.html
Nytimes
Google Says Criminal Hackers Used A.I. to Find a Major Software Flaw
The company said that it had identified, for the first time, hackers using artificial intelligence to discover an unknown bug. The attempted attack represents “a taste of what’s to come,” one expert said.