The July Model Wave Is Not a Race You Need to Win
Three frontier launches. Two weeks. One bad habit. The habit is crowning a winner from a press release. Claude Sonnet 5 on June 30. OpenAI's GPT-5.6 f…
Testing & QA
⚑ Report a ProblemLatest Testing & QA news from Tech News
All topics
agents
ai
api
architecture
automation
aws
backend
beginners
career
claude
cybersecurity
database
devchallenge
devops
javascript
llm
machinelearning
mcp
opensource
performance
productivity
programming
python
security
showdev
softwareengineering
testing
tutorial
typescript
webdev
Anthropic deleted 80% of Claude Code's system prompt. No regression.
Anthropic just published something that should make every developer building on Claude rethink their context engineering. For Claude Opus 5 and Fable …
I Grepped My Own Claude Code Logs and Found the Hidden Tag Anthropic Never Shows You
I was mid-conversation with Claude Code, asking it to help draft a blog post, when a literal <ip_reminder> tag showed up pasted into my own mess…
claude-security Beta: Cost, Output, Accuracy — 189 Agents, 2 Hours, Measured
What this article covers : A measured run of Anthropic's official security-scanning plugin claude-security (beta). What the tool does, how long it tak…
Claude Cowork on Windows and Linux: How to Run It in 2026
You bought a Max subscription on the strength of a Claude Cowork demo: a desktop agent quietly working through a folder of spreadsheets, drafting a de…
Where are YC founders now? OpenAI and Anthropic, mostly!
The Concentric Evolution of Y Combinator Alumni: From Generalist SaaS to Frontier AI The current landscape of the artificial intelligence industry is …
Ведущие ИИ-лабы естественным образом пришли к формату общей «выставки» для своих главных запусков
Anthropic, OpenAI, SpaceXAI и компания Цукерберга почти одновременно выкатили новые модели. Как будто съехались на конференцию, только виртуально. Чит…
Сравниваем LLM, 12 тестов для Claude Fable 5 и GPT 5.5 Pro
В прошлой статье мы устроили большое сравнение Opus 4.8, GPT 5.5 и Gemini 3.1 Pro и оговорились: GPT 5.5 Pro в том матче не участвовала, потому что ее…
Fable 5 из подвала, Sonnet 5 в дефолт. Что Antropic сделал за 5 дней
Утро четверга, в шапке Claude Code «Sonnet 5» — а я не переключал. Полез разбираться, что Antropic успел за 5 дней. Вернули Файбл 5 после 19 дней в по…
Anthropic just published a jailbreak severity scale. Here's what it means.
Anthropic has re-deployed Fable 5 and used the moment to publish two things that matter: a precise breakdown of what their cybersecurity classifiers w…
Claude Code was quietly fingerprinting requests through a hidden mark in the date
A reverse-engineer discovered that Anthropic's coding tool, Claude Code, embeds a hidden tracking mark in the system prompt it sends to its AI model. …
The Nobel Laureate Who Joined Anthropic Mid-Crisis
John Jumper announced on June 19 that he is leaving Google DeepMind to join Anthropic. He shared the 2024 Nobel Prize in Chemistry for protein structu…
This Week in AI: Claude Goes Dark, SpaceX Buys Cursor for $60B
This Week in AI: June 12–18, 2026 Six days. That's how long two of Anthropic's most capable models have been offline because of a single letter from t…
Anthropic’s Fable/Mythos shutdown is the first real model export-control shock
Anthropic’s Fable/Mythos shutdown is the first real model export-control shock The important AI story this week is not just that Anthropic launched bi…
Claude Design: What It Is, Where It Fits, and When to Skip It
Claude Design is a tool from Anthropic Labs that turns a conversation into editable visual work: prototypes, slide decks, one-pagers, mockups, landing…
AI news today: agent pricing, rare disease diagnosis, and China's local-model push
Today's AI news is not one of those neat, one-company launch days. It is messier than that. OpenAI is pushing AI into rare disease diagnosis. Anthropi…
Opus оркеструет, DeepSeek V4 пишет код: как собрать связку внутри Claude Code и сэкономить деньги
Однажды я открыл биллинг и просто посмотрел, на что уходят токены. Не на «подумать над архитектурой». А на переименование пер…
The Paradox of Power: Why Anthropic Released and Then Restricted Claude Fable 5
On June 9, 2026, Anthropic released Claude Fable 5, which was described as the most capable AI model publicly available at the time. Within 72 hours, …
Anthropic’s Claude in 2026: When Frontier AI Stopped Being Just Software
In 2026, Claude stopped looking like a normal AI product and started looking like infrastructure. Anthropic’s latest models are no longer interesting …
MCP Apps vs OpenAI Apps SDK: are they competing standards?
If you have started building tools for AI chat hosts, you have probably hit the same fork in the road. There are two ways to add a user interface to a…
Токен-оптимизация агентов: на что уходит контекстное окно MCP
Чем больше задач берёт на себя агент, тем чаще он упирается не в качество модели, а в контекстное окно: туда нужно уместить инструкции, историю диалог…
Claude Opus 4.8 shipped today. Here's the upgrade decision tree the announcement skipped — and three workloads that should stay on 4.7.
The 30-second version Anthropic shipped Claude Opus 4.8 a few hours ago. Every benchmark on the announcement page is up: SWE-bench Verified, GPQA, MAT…
ИИ уже пишет 80% кода Anthropic. Самое тревожное спрятано в цифре, которую подают как успех
Anthropic отчиталась, что больше 80% её кода теперь пишет Claude, — а её же автоматический проверяющий ловит лишь треть прошлых ошибок, то есть две тр…
Two agent skills hit GitHub trending the same week. Skills are becoming the new packages, and the dependency graph nobody is managing will bite by Q4.
The signal hidden in this week's GitHub trending Two agent-shaped repositories cracked the daily GitHub trending board this week. The first is mvanhor…
Claude Opus 4.8 shipped this week. The buried story is your migration cadence — your agent fleet won't survive the next four months without a refactor.
The benchmark is the wrong story Anthropic shipped Claude Opus 4.8 this week. You probably saw the announcement post on Tuesday, the swarm of benchmar…
Claude Opus 4.8 shipped today. Here is what the launch post does not say about why your agents will feel different tomorrow.
Claude Opus 4.8 shipped today. The benchmarks are a distraction — here is what actually changes about how your agents run tomorrow. Anthropic announce…
Opus 4.8 ships Dynamic Workflows — hundreds of parallel subagents per session. Read this before you wire it into prod.
Opus 4.8 ships Dynamic Workflows — hundreds of parallel subagents per session. Read this before you wire it into prod. Anthropic's Opus 4.8 announceme…
중국 암시장에서 Claude 10%에 판매 — 모델 증류의 진짜 위협
중국 암시장이 클로드를 10%에 팔고 있다, 그런데 앤트로픽이 정작 두려워하는 것은 따로 있다 모델 증류의 진짜 공포는 '가격'이 아니라 '속도'다 — 당신의 AI가 이미 모조품일 수 있다는 이야기 TL;DR : 중국 암시장에서 앤트로픽의 Claude가 원가의 10% …
AI обнулил benchmark и пытался шантажировать инженера. И почему это решаемо
Топовые AI-модели с 95% на SWE-bench показывают 0% и 3% на ProgramBench бенчмарке, где задачи специально не пересекаются с обучающей выборкой. Не «упа…
Shipped: first MCP server for ISO 10012:2026 measurement uncertainty
Spent 48 hours building a Model Context Protocol server for ISO 10012:2026 measurement uncertainty. 10 tools. JCGM 100:2008 plus 101:2008 compliant. L…