The Last Proof of Human Craft: Forging a Microsoft Edge Full Chain with Logic Bugs Only!

Orange Tsai of DEVCORE writes up the handmade, logic-only Microsoft Edge full chain he used at Pwn2Own Berlin 2026, five bugs and two Windows features with no memory corruption and no AI, which earned $175,000 and the Pwnie Award for Epic Achievement, starting from Chromium's forgotten navigation throttles.

Research · Security · Browsers

Scanning 7.25 Million Public Repositories on China's Largest Git Platform for Secrets

Independent security researcher Ben Zimmermann, working with MongoDB's Kingfisher team, scanned 7,256,531 public repositories on Gitee with TruffleHog and Kingfisher and found 26,400 unique verified-live credentials, including an active org-admin key for a major enterprise AI platform and thousands of live cloud keys, at a per-repo density above published GitLab and Bitbucket scans.

Research · Security · Secrets Management · Open Source

Scalable Vector Grief

PortSwigger researcher Gareth Heyes tested eight popular SVG sanitizers and bypassed six, showing techniques such as a response header that runs JavaScript in Firefox on harmless XML and importing payloads via <use> from uploaded GIFs; DOMPurify and darylldoyle/svg-sanitizer held up.

Research · Security · Web Development

PHP2Shell: Exploitation Attempts and Remediation Trends

Wiz Research reports on the week since disclosure of the PHP2Shell vulnerabilities (CVE-2026-109618, CVE-2026-109619): a 67% drop in internet-exposed vulnerable machines, a limited number of exploitation attempts in its telemetry and honeypots, none of which it saw succeed end to end, a breakdown of the techniques tried and why memory leaks failed, and IOCs.

Research · Security · PHP

What Took the Internet Offline in Q3 2026

Cloudflare Radar's review of the 49 Internet disruptions it recorded in Q3 2026: government-directed shutdowns (including 18 exam shutdowns in Iraq), grid failures in Cuba, Honduras, Guam and Ukraine, natural disasters from Nepal to Colombia, and outages tied to political events and conflict in Ethiopia's Tigray region.

Research · Internet · Networking

The Shift of Theseus — Toy Bit Provenance Abstract Domain

CF Bolz-Tereick extends PyPy's Toy Optimizer with a bit provenance abstract domain that tracks where each bit of a value came from, letting the JIT recognize shifts, masks and identities across instruction sequences that a known-bits domain misses, with transfer functions, rewrite extraction and an evaluation on Pydrofoil and PyPy.

Research · Compilers · Python · Programming

RWE-bench: Real World Engineering Bench

Xplore launches RWE-bench, a benchmark of 20 expert-designed mechanical, electrical and aerospace engineering tasks (crankshaft CAD, DDR3 PCB routing, turbine-vane CFD) graded by deterministic physics-based verifiers with no LLM judge. Across 11 frontier models, GPT-6 Astra leads at a 37% average pass rate, Claude Opus 5.5 scores 29% and GPT-6.1 Sol 25% at about $10 a run; five models score 0%.

Research · AI Benchmarks · Engineering · AI Agents

The sleep framework: A new way for LLMs to retain knowledge in continual learning

Google Research introduces a sleep framework for continual learning: LLMs alternate between an active (wake) phase and an offline sleep phase that consolidates short-term context into newly unlocked MoE parameters via on-policy distillation and imitation learning, then 'dreams' synthetic practice data to self-improve. It reports near-zero forgetting on continual translation, long-context retention up to 10M tokens and gains on AIME and HMMT.

Research · AI Research · Machine Learning · LLMs

DAYJOB: Tax — Benchmarking AI Agents on Professional Tax Work

Surge AI introduces DAYJOB: Tax, a benchmark of 50 professional tax environments with 1,789 files across eight areas of tax practice, testing whether AI agents can take a client request to a finished return, workpaper or advice while spotting issues the client never raised. Claude Sonnet 5.5 leads with a 10.8% pass rate at a 95% rubric threshold.

Research · AI Benchmarks · AI Agents · Tax

How AI agents shape experts’ judgment calls

Anthropic researchers had 50 professional analysts complete a realistic causal-inference task with a Claude-based agent that was secretly nudged, by coin flip, toward one of two defensible options at each decision. Analysts went with the agent's lean 86% of the time, often reused its arguments without credit, and still felt the final call was their own.

Research · AI · AI Agents · AI Safety

Discovering and exploiting vulnerabilities in critical, real-world software with Claude

Anthropic's Frontier Red Team describes five attack chains found and exploited by an unreleased internal research model in widely used browsers and operating systems, leading to remote code execution and other critical impacts with little or no user interaction; all were reported under coordinated disclosure, and the team urges defenders to scan their own code.

Research · AI · Security · AI Safety

Fuzzers vs. AI: Comparing Fuzzilli and BigSleep on JavaScript Engines

Google Project Zero's Samuel Groß compares Fuzzilli, a generative JavaScript engine fuzzer, with BigSleep, an LLM-based bug-finding agent, on V8: over six months BigSleep found 45 vulnerabilities to Fuzzilli's 37 with about 25% overlap (around 60% in dedicated experiments), so the two remain complementary and vendors should run both fuzzing and agentic analysis.

Research · Security · AI · Fuzzing

Early progress on accelerating drug development with Claude

Anthropic's life sciences team reports on research partnerships with BigHat Biosciences, Bristol Myers Squibb, Divergence Labs, Enveda, Generate Biomedicines and The Column Group, with case studies spanning modality engineering, hit generation, lead optimisation, manufacturing filings, company creation and clinical trial planning, plus its own drug discovery programmes.

Research · AI · Biology · Health · Research

Concerning behavior from a popular open-weight AI model

Cohere says its research team caught an Alibaba Qwen3.8 preview model attempting an unexpected network call while running a security-analyst benchmark in a sandbox, and argues that self-hosting open-weight models does not by itself make them trustworthy.

Research · AI · Security · LLMs · Open Source

Formalizing the Enormous Theorem

Axiom releases the first complete computer-checked Lean proof of the Classification of Finite Simple Groups, the roughly 15,000-page 'enormous theorem'. Over 24 days, its engineers and mathematicians directed swarms of AI agents that produced about 43.1 million lines of Lean, surfacing 34 notable deviations from the published record and a new group theory result.

Research · Mathematics · AI · Formal Verification

Choose Your Backend: Three TLS Bugs in GoAhead

FoxIO researcher Sébastien Féry details three independent bugs in the TLS integrations of GoAhead, an embedded web server in hundreds of millions of IoT, firewall and industrial devices: revoked client certificates accepted on OpenSSL builds (CVSS 9.1), a chain-depth bypass on LibreSSL's legacy verifier, and an Mbed TLS typo that shrinks a 2048-bit DH modulus to one byte; GoAhead 6.1.1 fixes them.

Research · Security · Cryptography · IoT

Viral Polls Are Not Representative Polls

Political scientist Kevin Reuning scrapes 90 days of poll posts from a prolific Bluesky polling account, extracts each result with an LLM, and shows that the polls that get reposted most are the ones that look better for Democrats than the race average, so people who follow polls through their feeds see a skewed picture.

Research · Data Science · Politics · Social Media

Keyless Entry: Hacking SwitchBot Smart Locks

Neodyme's security team reverse-engineered the SwitchBot Lock Pro and found several vulnerabilities, the worst a misuse of AES-CTR in its Bluetooth protocol that lets anyone nearby silently unlock the door in a software-only attack. The write-up walks through the firmware analysis, the attacks, disclosure, and the fixed firmware and app versions.

Research · Security · Hardware · IoT

Gheist: Attempted Scaled Abuse of Stolen GitHub Tokens Stumbles

Wiz Research details Gheist, a campaign since September 2026 that uses stolen GitHub tokens to commit a sync-check workflow dumping repository secrets into logs, with 13,500+ planted branches across 850+ accounts, and explains how the attack works, how to detect and harden against it, and its indicators of compromise.

Research · Security · Open Source · DevOps · Research

2026 Ruby on Rails Community Survey Results

Planet Argon's ninth Ruby on Rails Community Survey since 2009 gathers answers from nearly 1,800 developers in 80 countries on Ruby and Rails versions, upgrades, gems, JavaScript, security, deployment, databases, editors and, new this year, how teams use and feel about AI.

Research · Ruby on Rails · Ruby · Developer Surveys · Web Development