richardbarosky

Joined 6 years ago

Karma
918
Submissions
129
Member since
Aug 18
View on Hacker News

Recent submissions

Comment · 8 days ago
Tend to agree that for the strongest use cases, AI companies will prefer to sell the technique as a service rather than its raw output (e.g., their enterprise security services). For use cases where the output is less valuable, tokens are sold. If AI tokens were so magical in creating new value in developing software applications generally, they wouldn't be selling tokens directly. They'd hoard the tokens are use them to dominate SaaS software in any industry they want.

The same way as someone selling an expensive course in the stock market is signaling that they have more to gain by selling the course rather than taking their knowledge and making money in the stock market directly.

https://github.com/anthropics/defending-code-reference-harne...

Interestingly, it states:

"This repo is not maintained and is not accepting contributions.

Want a managed option? Anthropic offers Claude Security, a hosted product that finds and fixes vulnerabilities in your source code across multiple projects. Claude Security scans your repository for vulnerabilities, applies a multi-stage verification pipeline to reduce false positives, and lets you manage findings through their lifecycle: triage, fix validation, and rapid fix generation.

This repository is an open-source reference implementation based on general best practices for finding vulnerabilities using Claude. You can use it to build your own vulnerability finding pipeline, customize the logic, and it can be used with whatever access you have to Claude APIs (including Bedrock, Vertex, or Azure)."

Comment · 10 days ago
Comment · 10 days ago
Here's what I did at least.

I initially updated the following php.ini setting and verified that it updated via phpinfo(). I understand this to be only a possible mitigation (see below).

  session.upload_progress.enabled = Off
  file_uploads = Off
According to someone stating they are one of the PHP release managers:

  "INI settings alone won't protect you. You need to patch PHP."
https://www.reddit.com/r/PHP/comments/1xb2utu/comment/phyu5m...

I use Apache, and added an htaccess rewrite rule in my base directory to block all requests containing multi-part headers. Rewrite rules can also be added into Apache virtual host blocks. Accepting file uploads are not a critical function of the app in my case. This blocks multi-part form requests before they reach PHP.

  RewriteCond %{HTTP:Content-Type} multipart/form-data [NC]
  RewriteRule ^ - [F,L]
And verified the non-block vs. block case. https://freeimage.host/i/nOqumyN

I'm continuing to learn about the situation and may take the server offline as I learn more.

The writeup I found here:

https://jackalsyn.com/posts/php-up-2026.html

Content from a couple of my ChatGPT 6 Pro sessions about the issue:

https://chatgpt.com/share/6ad89ea3-b2f0-83ea-9c4c-2ecf5cc24b...

https://chatgpt.com/share/6ad8ad3d-3f6c-83e9-abf0-41da8a0d65...

  The vulnerability was rated 9.2 (critical) out of 10 under CVSS v4.0 by PHP developers,[1] while security firm watchTowr rated it a full 10.0.
https://en.wikipedia.org/wiki/PHP2Shell
Comment · 17 days ago
Would you mind sharing one or two of the top ones you saw? Definitely interested to check those out.
Comment · 1 month ago
Ha, I thought it was going to be the bet of trying to move as fast as possible with high volumes of AI generated code where the hope is that the AI itself will simply advance so rapidly as to make otherwise "maintainability challenged" code something that doesn't need to be considered.
Comment · 1 month ago
Writing about these things in public and putting yourself out there is greatly appreciated. Hats off for that!

However, as far as what's being pursued, it seems more like wantonly trying to ride a hype cycle without strongly questioning the end-to-end value of new software development approaches or vetting their immediate suitability.

Personally, I think it would be more sensible to take a few individuals or a smaller team(s) and do more isolated/skunkworks experimentation and adopt as justified based on what those people report/experience. The smaller group can adjust faster and iterate/advise the larger dev org about the good approaches/techniques/strategies, and avoid more broad damage/chaos for things that aren't that well thought out.

For more conservative AI use cases like adding to code review, writing low stakes PR summaries, or beefing up security checking, a more global, but still not off-the-rails, approach would be the kinds of things that would make more sense to push more broadly.

Comment · 1 month ago

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton