Mythos finds a curl vulnerability

Anthropic is positioning its gated Claude Mythos Preview model as offering a "striking leap" in many evaluation benchmarks over its predecessor, which at the time of launch was Opus 4.6.

Anthropic's Mythos showed that given enough inference, all bugs are shallow. But who pays for the inference? We benchmarked Claude Opus 4.6 against Corgea v1 and v2 to show why purpose-built scanner architecture beats raw model capability on precision, recall, cost, and speed.

A name identifies what object we want to access; an address identifies where the object is located. But there are subtleties to naming

AI did not invent new attacks. It billed the old ones monthly. Crypto is the only place we can count it, and the early signals are in.

GitHub and industry leaders warn of a massive surge in unvalidated AI vulnerability reporting, forcing bug bounty programs to tighten rules against the noise.

Google has patched an Android ADB bug in the May security patch set. If you have a Pixel phone you should already have the patches, and most other major manufacturers should be close behind. Unfort…

As LLMs supercharge CTF automation, the community must adapt to ensure learning incentives remain effective.

A new round of the weekly Security Affairs newsletter has arrived! Every week, the best security articles from Security Affairs.

Slashdot: News for nerds, stuff that matters. Timely news source for technology related news and B2B software reviews & comparisons.

aka 아델 #French 🇫🇷 #PHP / #JavaScript and #Java developer #Korean 🇰🇷 ancestry (but I don’t speak the language) Into #SmolWeb, #GeminiProtocol, #Smolnet, #LowTech #ArchLinux / #Debian user #Markdown 🇲⬇️ enthusiast Instance running #GoToSocial 🦥 en / fr :straightally:

i wear this chaos well

Your frontpage to the internet.

How I'm using Claude Opus 4.7 to find and report CVEs in popular Hex packages to make the BEAM ecosystem safer.

Personal website of Ankur Sethi, writer and independent frontend engineer living in Bangalore, India.

We are in Thrissur. We left at around 2 AM and reached by 11 AM, with two breaks. The drive was comfortable. The traffic was friendly this time. Echo is getting old, and Uma is more active now. So …

My Ministers will also introduce legislation to improve the country’s defences against cyber-security threats [Cyber Security and Resilience Bill].

As LLMs supercharge CTF automation, the community must adapt to ensure learning incentives remain effective.

AI comes for tech jobs. Security issues seem to be skyrocketing. General Catalyst Ad, Elon v OpenAI, Gamestop and eBay, Claude v Codex.

Teardown of TeamPCP's offensive framework that was briefly published on GitHub, Reddit AMA on a career in physical penetration testing, the end of "opaque defense": AI makes understanding defensive tool implementations easy

Weekly ThreatsDay Bulletin: supply chain attacks, fake support lures, AI tampering, data leaks, ransomware, and exploited flaws.

OpenAI seems to be trying for a much less scary vibe than Anthropic's.

How curl’s disclosure policy filtered an AI scanner’s findings at source

Slashdot: News for nerds, stuff that matters. Timely news source for technology related news and B2B software reviews & comparisons.

A computing-focused community centered around link aggregation and discussion.

cURL creator Daniel Stenberg says Anthropic's hyped Mythos bug-hunting model found only one confirmed low-severity vulnerability in cURL, plus a few non-security bugs, after he expected a much longer list. He argues Mythos may be useful, but not meaningfully beyond other modern AI code-analysis tool...

The tech giant says this is the first time it has identified such a case.

Something (slightly) less boring than nothing.

Daniel Stenberg has published a lengthy article on his thoughts on Anthropic's Mythos, which th [...]

Yesterday, I wrote about the massive increase in security issues being identified by AI. Daniel Stenberg, the author of curl, was a key part of that story. At first he was being overwhelmed by a torrent of “AI slop” reports. In the last few months, the reports have become almost all legitimate, but the pace hasn’t changed. Today, Stenberg posted about his recent experience with Claude Mythos.